Upload 13 files
Browse files- lora_8bit_adapter_raw/README.md +73 -0
- lora_8bit_adapter_raw/adapter_config.json +29 -0
- lora_8bit_adapter_raw/all_results.json +11 -0
- lora_8bit_adapter_raw/eval_results.json +7 -0
- lora_8bit_adapter_raw/special_tokens_map.json +35 -0
- lora_8bit_adapter_raw/tokenizer.model +3 -0
- lora_8bit_adapter_raw/tokenizer_config.json +74 -0
- lora_8bit_adapter_raw/train_results.json +7 -0
- lora_8bit_adapter_raw/trainer_log.jsonl +0 -0
- lora_8bit_adapter_raw/trainer_state.json +2727 -0
- lora_8bit_adapter_raw/training_args.bin +3 -0
- lora_8bit_adapter_raw/training_eval_loss.png +0 -0
- lora_8bit_adapter_raw/training_loss.png +0 -0
lora_8bit_adapter_raw/README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- llama-factory
|
| 6 |
+
- lora
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: content/drive/MyDrive/LLaMA-Factory/saves/Yi-6B-Chat/lora/lora_8bit_20w
|
| 9 |
+
model-index:
|
| 10 |
+
- name: oncology
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# oncology
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [content/drive/MyDrive/LLaMA-Factory/saves/Yi-6B-Chat/lora/lora_8bit_20w](https://huggingface.co/content/drive/MyDrive/LLaMA-Factory/saves/Yi-6B-Chat/lora/lora_8bit_20w) on the oncology_train_raw dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.6397
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0003
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- gradient_accumulation_steps: 4
|
| 45 |
+
- total_train_batch_size: 16
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_steps: 200
|
| 49 |
+
- num_epochs: 3.0
|
| 50 |
+
- mixed_precision_training: Native AMP
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 56 |
+
| 1.7882 | 0.32 | 200 | 1.7318 |
|
| 57 |
+
| 1.604 | 0.64 | 400 | 1.7021 |
|
| 58 |
+
| 1.706 | 0.96 | 600 | 1.6814 |
|
| 59 |
+
| 1.5361 | 1.28 | 800 | 1.6756 |
|
| 60 |
+
| 1.5625 | 1.6 | 1000 | 1.6565 |
|
| 61 |
+
| 1.606 | 1.91 | 1200 | 1.6396 |
|
| 62 |
+
| 1.4462 | 2.23 | 1400 | 1.6514 |
|
| 63 |
+
| 1.3236 | 2.55 | 1600 | 1.6478 |
|
| 64 |
+
| 1.4728 | 2.87 | 1800 | 1.6453 |
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
### Framework versions
|
| 68 |
+
|
| 69 |
+
- PEFT 0.10.0
|
| 70 |
+
- Transformers 4.38.2
|
| 71 |
+
- Pytorch 2.2.1+cu121
|
| 72 |
+
- Datasets 2.18.0
|
| 73 |
+
- Tokenizers 0.15.2
|
lora_8bit_adapter_raw/adapter_config.json
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "content/drive/MyDrive/LLaMA-Factory/saves/Yi-6B-Chat/lora/lora_8bit_20w",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layer_replication": null,
|
| 10 |
+
"layers_pattern": null,
|
| 11 |
+
"layers_to_transform": null,
|
| 12 |
+
"loftq_config": {},
|
| 13 |
+
"lora_alpha": 64,
|
| 14 |
+
"lora_dropout": 0.1,
|
| 15 |
+
"megatron_config": null,
|
| 16 |
+
"megatron_core": "megatron.core",
|
| 17 |
+
"modules_to_save": null,
|
| 18 |
+
"peft_type": "LORA",
|
| 19 |
+
"r": 32,
|
| 20 |
+
"rank_pattern": {},
|
| 21 |
+
"revision": null,
|
| 22 |
+
"target_modules": [
|
| 23 |
+
"v_proj",
|
| 24 |
+
"q_proj"
|
| 25 |
+
],
|
| 26 |
+
"task_type": "CAUSAL_LM",
|
| 27 |
+
"use_dora": false,
|
| 28 |
+
"use_rslora": false
|
| 29 |
+
}
|
lora_8bit_adapter_raw/all_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 1.639704942703247,
|
| 4 |
+
"eval_runtime": 54.9986,
|
| 5 |
+
"eval_samples_per_second": 8.8,
|
| 6 |
+
"eval_steps_per_second": 2.2,
|
| 7 |
+
"train_loss": 1.5555958059776072,
|
| 8 |
+
"train_runtime": 10085.6682,
|
| 9 |
+
"train_samples_per_second": 2.982,
|
| 10 |
+
"train_steps_per_second": 0.186
|
| 11 |
+
}
|
lora_8bit_adapter_raw/eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 1.639704942703247,
|
| 4 |
+
"eval_runtime": 54.9986,
|
| 5 |
+
"eval_samples_per_second": 8.8,
|
| 6 |
+
"eval_steps_per_second": 2.2
|
| 7 |
+
}
|
lora_8bit_adapter_raw/special_tokens_map.json
ADDED
|
@@ -0,0 +1,35 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>",
|
| 5 |
+
"<|im_sep|>"
|
| 6 |
+
],
|
| 7 |
+
"bos_token": {
|
| 8 |
+
"content": "<|startoftext|>",
|
| 9 |
+
"lstrip": false,
|
| 10 |
+
"normalized": true,
|
| 11 |
+
"rstrip": false,
|
| 12 |
+
"single_word": false
|
| 13 |
+
},
|
| 14 |
+
"eos_token": {
|
| 15 |
+
"content": "<|im_end|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false
|
| 20 |
+
},
|
| 21 |
+
"pad_token": {
|
| 22 |
+
"content": "<unk>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": true,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false
|
| 27 |
+
},
|
| 28 |
+
"unk_token": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false
|
| 34 |
+
}
|
| 35 |
+
}
|
lora_8bit_adapter_raw/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:386c49cf943d71aa110361135338c50e38beeff0a66593480421f37b319e1a39
|
| 3 |
+
size 1033105
|
lora_8bit_adapter_raw/tokenizer_config.json
ADDED
|
@@ -0,0 +1,74 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": true,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<|startoftext|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": true,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "<|endoftext|>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": true,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
},
|
| 30 |
+
"6": {
|
| 31 |
+
"content": "<|im_start|>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false,
|
| 36 |
+
"special": true
|
| 37 |
+
},
|
| 38 |
+
"7": {
|
| 39 |
+
"content": "<|im_end|>",
|
| 40 |
+
"lstrip": false,
|
| 41 |
+
"normalized": false,
|
| 42 |
+
"rstrip": false,
|
| 43 |
+
"single_word": false,
|
| 44 |
+
"special": true
|
| 45 |
+
},
|
| 46 |
+
"8": {
|
| 47 |
+
"content": "<|im_sep|>",
|
| 48 |
+
"lstrip": false,
|
| 49 |
+
"normalized": false,
|
| 50 |
+
"rstrip": false,
|
| 51 |
+
"single_word": false,
|
| 52 |
+
"special": true
|
| 53 |
+
}
|
| 54 |
+
},
|
| 55 |
+
"additional_special_tokens": [
|
| 56 |
+
"<|im_start|>",
|
| 57 |
+
"<|im_end|>",
|
| 58 |
+
"<|im_sep|>"
|
| 59 |
+
],
|
| 60 |
+
"bos_token": "<|startoftext|>",
|
| 61 |
+
"chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|im_start|>user\\n' + content + '<|im_end|>\\n<|im_start|>assistant\\n' }}{% elif message['role'] == 'assistant' %}{{ content + '<|im_end|>' + '\\n' }}{% endif %}{% endfor %}",
|
| 62 |
+
"clean_up_tokenization_spaces": false,
|
| 63 |
+
"eos_token": "<|im_end|>",
|
| 64 |
+
"legacy": true,
|
| 65 |
+
"model_max_length": 4096,
|
| 66 |
+
"pad_token": "<unk>",
|
| 67 |
+
"padding_side": "right",
|
| 68 |
+
"sp_model_kwargs": {},
|
| 69 |
+
"spaces_between_special_tokens": false,
|
| 70 |
+
"split_special_tokens": false,
|
| 71 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 72 |
+
"unk_token": "<unk>",
|
| 73 |
+
"use_default_system_prompt": true
|
| 74 |
+
}
|
lora_8bit_adapter_raw/train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.5555958059776072,
|
| 4 |
+
"train_runtime": 10085.6682,
|
| 5 |
+
"train_samples_per_second": 2.982,
|
| 6 |
+
"train_steps_per_second": 0.186
|
| 7 |
+
}
|
lora_8bit_adapter_raw/trainer_log.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
lora_8bit_adapter_raw/trainer_state.json
ADDED
|
@@ -0,0 +1,2727 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.6396043300628662,
|
| 3 |
+
"best_model_checkpoint": "content/lora_8bit_onco_raw/oncology/checkpoint-1200",
|
| 4 |
+
"epoch": 2.9964100518548067,
|
| 5 |
+
"eval_steps": 200,
|
| 6 |
+
"global_step": 1878,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 0.7856405973434448,
|
| 14 |
+
"learning_rate": 7.499999999999999e-06,
|
| 15 |
+
"loss": 1.9839,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"grad_norm": 0.7171689867973328,
|
| 21 |
+
"learning_rate": 1.4999999999999999e-05,
|
| 22 |
+
"loss": 1.8072,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02,
|
| 27 |
+
"grad_norm": 0.7502748966217041,
|
| 28 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 29 |
+
"loss": 1.916,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.03,
|
| 34 |
+
"grad_norm": 0.6175706386566162,
|
| 35 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 36 |
+
"loss": 1.8292,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.04,
|
| 41 |
+
"grad_norm": 0.4654443562030792,
|
| 42 |
+
"learning_rate": 3.75e-05,
|
| 43 |
+
"loss": 1.8564,
|
| 44 |
+
"step": 25
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05,
|
| 48 |
+
"grad_norm": 0.5197769999504089,
|
| 49 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 50 |
+
"loss": 1.9732,
|
| 51 |
+
"step": 30
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.06,
|
| 55 |
+
"grad_norm": 0.6166006922721863,
|
| 56 |
+
"learning_rate": 5.2499999999999995e-05,
|
| 57 |
+
"loss": 1.8221,
|
| 58 |
+
"step": 35
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.06,
|
| 62 |
+
"grad_norm": 0.5110687613487244,
|
| 63 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 64 |
+
"loss": 1.5955,
|
| 65 |
+
"step": 40
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.07,
|
| 69 |
+
"grad_norm": 0.6438449025154114,
|
| 70 |
+
"learning_rate": 6.75e-05,
|
| 71 |
+
"loss": 1.7571,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.08,
|
| 76 |
+
"grad_norm": 0.4936439096927643,
|
| 77 |
+
"learning_rate": 7.5e-05,
|
| 78 |
+
"loss": 1.6779,
|
| 79 |
+
"step": 50
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.09,
|
| 83 |
+
"grad_norm": 0.554935872554779,
|
| 84 |
+
"learning_rate": 8.25e-05,
|
| 85 |
+
"loss": 1.74,
|
| 86 |
+
"step": 55
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1,
|
| 90 |
+
"grad_norm": 0.49913492798805237,
|
| 91 |
+
"learning_rate": 8.999999999999999e-05,
|
| 92 |
+
"loss": 1.8844,
|
| 93 |
+
"step": 60
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1,
|
| 97 |
+
"grad_norm": 0.5313754081726074,
|
| 98 |
+
"learning_rate": 9.75e-05,
|
| 99 |
+
"loss": 1.6513,
|
| 100 |
+
"step": 65
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.11,
|
| 104 |
+
"grad_norm": 0.5471957921981812,
|
| 105 |
+
"learning_rate": 0.00010499999999999999,
|
| 106 |
+
"loss": 1.7246,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.12,
|
| 111 |
+
"grad_norm": 0.6044861674308777,
|
| 112 |
+
"learning_rate": 0.0001125,
|
| 113 |
+
"loss": 1.8523,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.13,
|
| 118 |
+
"grad_norm": 0.5198370814323425,
|
| 119 |
+
"learning_rate": 0.00011999999999999999,
|
| 120 |
+
"loss": 1.7499,
|
| 121 |
+
"step": 80
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.14,
|
| 125 |
+
"grad_norm": 0.5604796409606934,
|
| 126 |
+
"learning_rate": 0.00012749999999999998,
|
| 127 |
+
"loss": 1.7664,
|
| 128 |
+
"step": 85
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.14,
|
| 132 |
+
"grad_norm": 0.5335713624954224,
|
| 133 |
+
"learning_rate": 0.000135,
|
| 134 |
+
"loss": 1.666,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.15,
|
| 139 |
+
"grad_norm": 0.6011459827423096,
|
| 140 |
+
"learning_rate": 0.0001425,
|
| 141 |
+
"loss": 1.8222,
|
| 142 |
+
"step": 95
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.16,
|
| 146 |
+
"grad_norm": 0.6249136924743652,
|
| 147 |
+
"learning_rate": 0.00015,
|
| 148 |
+
"loss": 1.7571,
|
| 149 |
+
"step": 100
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.17,
|
| 153 |
+
"grad_norm": 0.5300886034965515,
|
| 154 |
+
"learning_rate": 0.00015749999999999998,
|
| 155 |
+
"loss": 1.7534,
|
| 156 |
+
"step": 105
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.18,
|
| 160 |
+
"grad_norm": 0.5625594854354858,
|
| 161 |
+
"learning_rate": 0.000165,
|
| 162 |
+
"loss": 1.6312,
|
| 163 |
+
"step": 110
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.18,
|
| 167 |
+
"grad_norm": 0.48619601130485535,
|
| 168 |
+
"learning_rate": 0.00017249999999999996,
|
| 169 |
+
"loss": 1.7695,
|
| 170 |
+
"step": 115
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.19,
|
| 174 |
+
"grad_norm": 0.5377641320228577,
|
| 175 |
+
"learning_rate": 0.00017999999999999998,
|
| 176 |
+
"loss": 1.7723,
|
| 177 |
+
"step": 120
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.2,
|
| 181 |
+
"grad_norm": 0.5358831882476807,
|
| 182 |
+
"learning_rate": 0.00018749999999999998,
|
| 183 |
+
"loss": 1.7541,
|
| 184 |
+
"step": 125
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.21,
|
| 188 |
+
"grad_norm": 0.5402532815933228,
|
| 189 |
+
"learning_rate": 0.000195,
|
| 190 |
+
"loss": 1.6796,
|
| 191 |
+
"step": 130
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.22,
|
| 195 |
+
"grad_norm": 0.5598291754722595,
|
| 196 |
+
"learning_rate": 0.0002025,
|
| 197 |
+
"loss": 1.6744,
|
| 198 |
+
"step": 135
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.22,
|
| 202 |
+
"grad_norm": 0.5184920430183411,
|
| 203 |
+
"learning_rate": 0.00020999999999999998,
|
| 204 |
+
"loss": 1.7768,
|
| 205 |
+
"step": 140
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.23,
|
| 209 |
+
"grad_norm": 0.5164936780929565,
|
| 210 |
+
"learning_rate": 0.00021749999999999997,
|
| 211 |
+
"loss": 1.6877,
|
| 212 |
+
"step": 145
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.24,
|
| 216 |
+
"grad_norm": 0.4958161413669586,
|
| 217 |
+
"learning_rate": 0.000225,
|
| 218 |
+
"loss": 1.8754,
|
| 219 |
+
"step": 150
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.25,
|
| 223 |
+
"grad_norm": 0.564961314201355,
|
| 224 |
+
"learning_rate": 0.00023249999999999999,
|
| 225 |
+
"loss": 1.6324,
|
| 226 |
+
"step": 155
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.26,
|
| 230 |
+
"grad_norm": 0.8469408750534058,
|
| 231 |
+
"learning_rate": 0.00023999999999999998,
|
| 232 |
+
"loss": 1.7146,
|
| 233 |
+
"step": 160
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.26,
|
| 237 |
+
"grad_norm": 0.6041777729988098,
|
| 238 |
+
"learning_rate": 0.00024749999999999994,
|
| 239 |
+
"loss": 1.6416,
|
| 240 |
+
"step": 165
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.27,
|
| 244 |
+
"grad_norm": 0.5473640561103821,
|
| 245 |
+
"learning_rate": 0.00025499999999999996,
|
| 246 |
+
"loss": 1.6726,
|
| 247 |
+
"step": 170
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.28,
|
| 251 |
+
"grad_norm": 0.6132262349128723,
|
| 252 |
+
"learning_rate": 0.0002625,
|
| 253 |
+
"loss": 1.7383,
|
| 254 |
+
"step": 175
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.29,
|
| 258 |
+
"grad_norm": 0.49731534719467163,
|
| 259 |
+
"learning_rate": 0.00027,
|
| 260 |
+
"loss": 1.6531,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.3,
|
| 265 |
+
"grad_norm": 0.5386232733726501,
|
| 266 |
+
"learning_rate": 0.00027749999999999997,
|
| 267 |
+
"loss": 1.7335,
|
| 268 |
+
"step": 185
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.3,
|
| 272 |
+
"grad_norm": 0.5605046153068542,
|
| 273 |
+
"learning_rate": 0.000285,
|
| 274 |
+
"loss": 1.7373,
|
| 275 |
+
"step": 190
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.31,
|
| 279 |
+
"grad_norm": 0.7236664295196533,
|
| 280 |
+
"learning_rate": 0.00029249999999999995,
|
| 281 |
+
"loss": 1.5766,
|
| 282 |
+
"step": 195
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.32,
|
| 286 |
+
"grad_norm": 0.5735570788383484,
|
| 287 |
+
"learning_rate": 0.0003,
|
| 288 |
+
"loss": 1.7882,
|
| 289 |
+
"step": 200
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.32,
|
| 293 |
+
"eval_loss": 1.7317848205566406,
|
| 294 |
+
"eval_runtime": 53.4548,
|
| 295 |
+
"eval_samples_per_second": 9.054,
|
| 296 |
+
"eval_steps_per_second": 2.264,
|
| 297 |
+
"step": 200
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.33,
|
| 301 |
+
"grad_norm": 0.48291996121406555,
|
| 302 |
+
"learning_rate": 0.0002999934277521501,
|
| 303 |
+
"loss": 1.6244,
|
| 304 |
+
"step": 205
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.34,
|
| 308 |
+
"grad_norm": 0.5112442970275879,
|
| 309 |
+
"learning_rate": 0.00029997371158452626,
|
| 310 |
+
"loss": 1.6932,
|
| 311 |
+
"step": 210
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.34,
|
| 315 |
+
"grad_norm": 0.5557854771614075,
|
| 316 |
+
"learning_rate": 0.0002999408532248557,
|
| 317 |
+
"loss": 1.6655,
|
| 318 |
+
"step": 215
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.35,
|
| 322 |
+
"grad_norm": 0.477541446685791,
|
| 323 |
+
"learning_rate": 0.00029989485555251554,
|
| 324 |
+
"loss": 1.6773,
|
| 325 |
+
"step": 220
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.36,
|
| 329 |
+
"grad_norm": 0.45816484093666077,
|
| 330 |
+
"learning_rate": 0.00029983572259828045,
|
| 331 |
+
"loss": 1.5355,
|
| 332 |
+
"step": 225
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.37,
|
| 336 |
+
"grad_norm": 0.5112146735191345,
|
| 337 |
+
"learning_rate": 0.00029976345954396954,
|
| 338 |
+
"loss": 1.6489,
|
| 339 |
+
"step": 230
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.37,
|
| 343 |
+
"grad_norm": 0.5796986222267151,
|
| 344 |
+
"learning_rate": 0.0002996780727219923,
|
| 345 |
+
"loss": 1.6708,
|
| 346 |
+
"step": 235
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.38,
|
| 350 |
+
"grad_norm": 0.4717162251472473,
|
| 351 |
+
"learning_rate": 0.0002995795696147933,
|
| 352 |
+
"loss": 1.7605,
|
| 353 |
+
"step": 240
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.39,
|
| 357 |
+
"grad_norm": 0.5081738233566284,
|
| 358 |
+
"learning_rate": 0.00029946795885419714,
|
| 359 |
+
"loss": 1.7257,
|
| 360 |
+
"step": 245
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.4,
|
| 364 |
+
"grad_norm": 0.6322786808013916,
|
| 365 |
+
"learning_rate": 0.0002993432502206515,
|
| 366 |
+
"loss": 1.7894,
|
| 367 |
+
"step": 250
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.41,
|
| 371 |
+
"grad_norm": 0.626715898513794,
|
| 372 |
+
"learning_rate": 0.00029920545464237033,
|
| 373 |
+
"loss": 1.6251,
|
| 374 |
+
"step": 255
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.41,
|
| 378 |
+
"grad_norm": 0.5590308308601379,
|
| 379 |
+
"learning_rate": 0.0002990545841943763,
|
| 380 |
+
"loss": 1.602,
|
| 381 |
+
"step": 260
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.42,
|
| 385 |
+
"grad_norm": 0.5137120485305786,
|
| 386 |
+
"learning_rate": 0.00029889065209744235,
|
| 387 |
+
"loss": 1.664,
|
| 388 |
+
"step": 265
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.43,
|
| 392 |
+
"grad_norm": 0.5325127840042114,
|
| 393 |
+
"learning_rate": 0.0002987136727169335,
|
| 394 |
+
"loss": 1.5928,
|
| 395 |
+
"step": 270
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.44,
|
| 399 |
+
"grad_norm": 0.5906115174293518,
|
| 400 |
+
"learning_rate": 0.0002985236615615478,
|
| 401 |
+
"loss": 1.5565,
|
| 402 |
+
"step": 275
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.45,
|
| 406 |
+
"grad_norm": 0.5913872718811035,
|
| 407 |
+
"learning_rate": 0.00029832063528195733,
|
| 408 |
+
"loss": 1.701,
|
| 409 |
+
"step": 280
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.45,
|
| 413 |
+
"grad_norm": 0.5970895886421204,
|
| 414 |
+
"learning_rate": 0.0002981046116693491,
|
| 415 |
+
"loss": 1.5652,
|
| 416 |
+
"step": 285
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.46,
|
| 420 |
+
"grad_norm": 0.5957712531089783,
|
| 421 |
+
"learning_rate": 0.00029787560965386614,
|
| 422 |
+
"loss": 1.6611,
|
| 423 |
+
"step": 290
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.47,
|
| 427 |
+
"grad_norm": 0.5931127071380615,
|
| 428 |
+
"learning_rate": 0.00029763364930294854,
|
| 429 |
+
"loss": 1.6456,
|
| 430 |
+
"step": 295
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.48,
|
| 434 |
+
"grad_norm": 0.5497797131538391,
|
| 435 |
+
"learning_rate": 0.00029737875181957486,
|
| 436 |
+
"loss": 1.5782,
|
| 437 |
+
"step": 300
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.49,
|
| 441 |
+
"grad_norm": 0.5486745238304138,
|
| 442 |
+
"learning_rate": 0.00029711093954040425,
|
| 443 |
+
"loss": 1.7476,
|
| 444 |
+
"step": 305
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.49,
|
| 448 |
+
"grad_norm": 0.5858426690101624,
|
| 449 |
+
"learning_rate": 0.0002968302359338191,
|
| 450 |
+
"loss": 1.6871,
|
| 451 |
+
"step": 310
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5,
|
| 455 |
+
"grad_norm": 0.5358626842498779,
|
| 456 |
+
"learning_rate": 0.0002965366655978683,
|
| 457 |
+
"loss": 1.8075,
|
| 458 |
+
"step": 315
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.51,
|
| 462 |
+
"grad_norm": 0.6227905750274658,
|
| 463 |
+
"learning_rate": 0.00029623025425811215,
|
| 464 |
+
"loss": 1.6059,
|
| 465 |
+
"step": 320
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.52,
|
| 469 |
+
"grad_norm": 0.5188565850257874,
|
| 470 |
+
"learning_rate": 0.0002959110287653674,
|
| 471 |
+
"loss": 1.7475,
|
| 472 |
+
"step": 325
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.53,
|
| 476 |
+
"grad_norm": 0.5319951772689819,
|
| 477 |
+
"learning_rate": 0.00029557901709335497,
|
| 478 |
+
"loss": 1.6401,
|
| 479 |
+
"step": 330
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.53,
|
| 483 |
+
"grad_norm": 0.5894659757614136,
|
| 484 |
+
"learning_rate": 0.00029523424833624806,
|
| 485 |
+
"loss": 1.7475,
|
| 486 |
+
"step": 335
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.54,
|
| 490 |
+
"grad_norm": 0.5921294093132019,
|
| 491 |
+
"learning_rate": 0.00029487675270612303,
|
| 492 |
+
"loss": 1.7581,
|
| 493 |
+
"step": 340
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.55,
|
| 497 |
+
"grad_norm": 0.528262734413147,
|
| 498 |
+
"learning_rate": 0.0002945065615303116,
|
| 499 |
+
"loss": 1.8347,
|
| 500 |
+
"step": 345
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.56,
|
| 504 |
+
"grad_norm": 0.6218762397766113,
|
| 505 |
+
"learning_rate": 0.000294123707248656,
|
| 506 |
+
"loss": 1.789,
|
| 507 |
+
"step": 350
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.57,
|
| 511 |
+
"grad_norm": 0.6058654189109802,
|
| 512 |
+
"learning_rate": 0.000293728223410666,
|
| 513 |
+
"loss": 1.646,
|
| 514 |
+
"step": 355
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.57,
|
| 518 |
+
"grad_norm": 0.5830885767936707,
|
| 519 |
+
"learning_rate": 0.00029332014467257884,
|
| 520 |
+
"loss": 1.7596,
|
| 521 |
+
"step": 360
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.58,
|
| 525 |
+
"grad_norm": 0.5829673409461975,
|
| 526 |
+
"learning_rate": 0.0002928995067943227,
|
| 527 |
+
"loss": 1.5936,
|
| 528 |
+
"step": 365
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.59,
|
| 532 |
+
"grad_norm": 0.569988489151001,
|
| 533 |
+
"learning_rate": 0.00029246634663638286,
|
| 534 |
+
"loss": 1.8244,
|
| 535 |
+
"step": 370
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6,
|
| 539 |
+
"grad_norm": 0.5579255819320679,
|
| 540 |
+
"learning_rate": 0.0002920207021565714,
|
| 541 |
+
"loss": 1.7428,
|
| 542 |
+
"step": 375
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.61,
|
| 546 |
+
"grad_norm": 0.5439159274101257,
|
| 547 |
+
"learning_rate": 0.00029156261240670145,
|
| 548 |
+
"loss": 1.6165,
|
| 549 |
+
"step": 380
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.61,
|
| 553 |
+
"grad_norm": 0.6219534873962402,
|
| 554 |
+
"learning_rate": 0.0002910921175291646,
|
| 555 |
+
"loss": 1.727,
|
| 556 |
+
"step": 385
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.62,
|
| 560 |
+
"grad_norm": 0.5160556435585022,
|
| 561 |
+
"learning_rate": 0.00029060925875341344,
|
| 562 |
+
"loss": 1.5996,
|
| 563 |
+
"step": 390
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.63,
|
| 567 |
+
"grad_norm": 0.6084945797920227,
|
| 568 |
+
"learning_rate": 0.0002901140783923487,
|
| 569 |
+
"loss": 1.6613,
|
| 570 |
+
"step": 395
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.64,
|
| 574 |
+
"grad_norm": 0.552727222442627,
|
| 575 |
+
"learning_rate": 0.00028960661983861124,
|
| 576 |
+
"loss": 1.604,
|
| 577 |
+
"step": 400
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.64,
|
| 581 |
+
"eval_loss": 1.7020632028579712,
|
| 582 |
+
"eval_runtime": 53.5236,
|
| 583 |
+
"eval_samples_per_second": 9.043,
|
| 584 |
+
"eval_steps_per_second": 2.261,
|
| 585 |
+
"step": 400
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.65,
|
| 589 |
+
"grad_norm": 0.6136419177055359,
|
| 590 |
+
"learning_rate": 0.00028908692756077957,
|
| 591 |
+
"loss": 1.6674,
|
| 592 |
+
"step": 405
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.65,
|
| 596 |
+
"grad_norm": 0.6204317212104797,
|
| 597 |
+
"learning_rate": 0.00028855504709947305,
|
| 598 |
+
"loss": 1.7643,
|
| 599 |
+
"step": 410
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 0.66,
|
| 603 |
+
"grad_norm": 0.5804699659347534,
|
| 604 |
+
"learning_rate": 0.0002880110250633613,
|
| 605 |
+
"loss": 1.5933,
|
| 606 |
+
"step": 415
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 0.67,
|
| 610 |
+
"grad_norm": 0.5574301481246948,
|
| 611 |
+
"learning_rate": 0.0002874549091250797,
|
| 612 |
+
"loss": 1.6314,
|
| 613 |
+
"step": 420
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 0.68,
|
| 617 |
+
"grad_norm": 0.666695773601532,
|
| 618 |
+
"learning_rate": 0.00028688674801705203,
|
| 619 |
+
"loss": 1.62,
|
| 620 |
+
"step": 425
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.69,
|
| 624 |
+
"grad_norm": 0.5769309997558594,
|
| 625 |
+
"learning_rate": 0.0002863065915272199,
|
| 626 |
+
"loss": 1.7478,
|
| 627 |
+
"step": 430
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.69,
|
| 631 |
+
"grad_norm": 0.5807488560676575,
|
| 632 |
+
"learning_rate": 0.0002857144904946799,
|
| 633 |
+
"loss": 1.7793,
|
| 634 |
+
"step": 435
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 0.7,
|
| 638 |
+
"grad_norm": 0.60151606798172,
|
| 639 |
+
"learning_rate": 0.0002851104968052284,
|
| 640 |
+
"loss": 1.6921,
|
| 641 |
+
"step": 440
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 0.71,
|
| 645 |
+
"grad_norm": 0.5778372883796692,
|
| 646 |
+
"learning_rate": 0.00028449466338681525,
|
| 647 |
+
"loss": 1.6779,
|
| 648 |
+
"step": 445
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 0.72,
|
| 652 |
+
"grad_norm": 0.6416473388671875,
|
| 653 |
+
"learning_rate": 0.00028386704420490513,
|
| 654 |
+
"loss": 1.6848,
|
| 655 |
+
"step": 450
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 0.73,
|
| 659 |
+
"grad_norm": 0.681058943271637,
|
| 660 |
+
"learning_rate": 0.0002832276942577491,
|
| 661 |
+
"loss": 1.7468,
|
| 662 |
+
"step": 455
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.73,
|
| 666 |
+
"grad_norm": 0.6174286007881165,
|
| 667 |
+
"learning_rate": 0.0002825766695715646,
|
| 668 |
+
"loss": 1.6839,
|
| 669 |
+
"step": 460
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.74,
|
| 673 |
+
"grad_norm": 0.742919385433197,
|
| 674 |
+
"learning_rate": 0.0002819140271956262,
|
| 675 |
+
"loss": 1.7846,
|
| 676 |
+
"step": 465
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.75,
|
| 680 |
+
"grad_norm": 0.744145929813385,
|
| 681 |
+
"learning_rate": 0.00028123982519726634,
|
| 682 |
+
"loss": 1.6862,
|
| 683 |
+
"step": 470
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.76,
|
| 687 |
+
"grad_norm": 0.590994656085968,
|
| 688 |
+
"learning_rate": 0.0002805541226567869,
|
| 689 |
+
"loss": 1.6675,
|
| 690 |
+
"step": 475
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 0.77,
|
| 694 |
+
"grad_norm": 0.6487185955047607,
|
| 695 |
+
"learning_rate": 0.0002798569796622817,
|
| 696 |
+
"loss": 1.5819,
|
| 697 |
+
"step": 480
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 0.77,
|
| 701 |
+
"grad_norm": 0.6046218276023865,
|
| 702 |
+
"learning_rate": 0.00027914845730437135,
|
| 703 |
+
"loss": 1.6375,
|
| 704 |
+
"step": 485
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.78,
|
| 708 |
+
"grad_norm": 0.6337038278579712,
|
| 709 |
+
"learning_rate": 0.00027842861767084996,
|
| 710 |
+
"loss": 1.8259,
|
| 711 |
+
"step": 490
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.79,
|
| 715 |
+
"grad_norm": 0.6214140057563782,
|
| 716 |
+
"learning_rate": 0.00027769752384124375,
|
| 717 |
+
"loss": 1.7048,
|
| 718 |
+
"step": 495
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.8,
|
| 722 |
+
"grad_norm": 0.646763026714325,
|
| 723 |
+
"learning_rate": 0.00027695523988128417,
|
| 724 |
+
"loss": 1.7394,
|
| 725 |
+
"step": 500
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 0.81,
|
| 729 |
+
"grad_norm": 0.5836499929428101,
|
| 730 |
+
"learning_rate": 0.0002762018308372934,
|
| 731 |
+
"loss": 1.6096,
|
| 732 |
+
"step": 505
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 0.81,
|
| 736 |
+
"grad_norm": 0.557185709476471,
|
| 737 |
+
"learning_rate": 0.0002754373627304842,
|
| 738 |
+
"loss": 1.6317,
|
| 739 |
+
"step": 510
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 0.82,
|
| 743 |
+
"grad_norm": 0.5877421498298645,
|
| 744 |
+
"learning_rate": 0.00027466190255117506,
|
| 745 |
+
"loss": 1.6814,
|
| 746 |
+
"step": 515
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.83,
|
| 750 |
+
"grad_norm": 0.6311076283454895,
|
| 751 |
+
"learning_rate": 0.00027387551825291907,
|
| 752 |
+
"loss": 1.7358,
|
| 753 |
+
"step": 520
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.84,
|
| 757 |
+
"grad_norm": 0.57215815782547,
|
| 758 |
+
"learning_rate": 0.0002730782787465499,
|
| 759 |
+
"loss": 1.5929,
|
| 760 |
+
"step": 525
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 0.85,
|
| 764 |
+
"grad_norm": 0.6365725994110107,
|
| 765 |
+
"learning_rate": 0.0002722702538941425,
|
| 766 |
+
"loss": 1.6938,
|
| 767 |
+
"step": 530
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 0.85,
|
| 771 |
+
"grad_norm": 0.6177056431770325,
|
| 772 |
+
"learning_rate": 0.0002714515145028916,
|
| 773 |
+
"loss": 1.6506,
|
| 774 |
+
"step": 535
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 0.86,
|
| 778 |
+
"grad_norm": 0.6062068343162537,
|
| 779 |
+
"learning_rate": 0.0002706221323189066,
|
| 780 |
+
"loss": 1.7848,
|
| 781 |
+
"step": 540
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 0.87,
|
| 785 |
+
"grad_norm": 0.6754643321037292,
|
| 786 |
+
"learning_rate": 0.0002697821800209244,
|
| 787 |
+
"loss": 1.6818,
|
| 788 |
+
"step": 545
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.88,
|
| 792 |
+
"grad_norm": 0.5303813219070435,
|
| 793 |
+
"learning_rate": 0.00026893173121394094,
|
| 794 |
+
"loss": 1.6557,
|
| 795 |
+
"step": 550
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.89,
|
| 799 |
+
"grad_norm": 0.560896635055542,
|
| 800 |
+
"learning_rate": 0.00026807086042276074,
|
| 801 |
+
"loss": 1.5825,
|
| 802 |
+
"step": 555
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 0.89,
|
| 806 |
+
"grad_norm": 0.5587555766105652,
|
| 807 |
+
"learning_rate": 0.00026719964308546647,
|
| 808 |
+
"loss": 1.8095,
|
| 809 |
+
"step": 560
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 0.9,
|
| 813 |
+
"grad_norm": 0.5372602343559265,
|
| 814 |
+
"learning_rate": 0.00026631815554680854,
|
| 815 |
+
"loss": 1.5673,
|
| 816 |
+
"step": 565
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 0.91,
|
| 820 |
+
"grad_norm": 0.7331911325454712,
|
| 821 |
+
"learning_rate": 0.0002654264750515146,
|
| 822 |
+
"loss": 1.6092,
|
| 823 |
+
"step": 570
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.92,
|
| 827 |
+
"grad_norm": 0.5914915800094604,
|
| 828 |
+
"learning_rate": 0.00026452467973752097,
|
| 829 |
+
"loss": 1.6226,
|
| 830 |
+
"step": 575
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.93,
|
| 834 |
+
"grad_norm": 0.6332432627677917,
|
| 835 |
+
"learning_rate": 0.0002636128486291251,
|
| 836 |
+
"loss": 1.687,
|
| 837 |
+
"step": 580
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.93,
|
| 841 |
+
"grad_norm": 0.5786017179489136,
|
| 842 |
+
"learning_rate": 0.00026269106163006093,
|
| 843 |
+
"loss": 1.7443,
|
| 844 |
+
"step": 585
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 0.94,
|
| 848 |
+
"grad_norm": 0.6086643934249878,
|
| 849 |
+
"learning_rate": 0.0002617593995164967,
|
| 850 |
+
"loss": 1.6546,
|
| 851 |
+
"step": 590
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 0.95,
|
| 855 |
+
"grad_norm": 0.6353456377983093,
|
| 856 |
+
"learning_rate": 0.00026081794392995676,
|
| 857 |
+
"loss": 1.6635,
|
| 858 |
+
"step": 595
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 0.96,
|
| 862 |
+
"grad_norm": 0.6383431553840637,
|
| 863 |
+
"learning_rate": 0.00025986677737016713,
|
| 864 |
+
"loss": 1.706,
|
| 865 |
+
"step": 600
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 0.96,
|
| 869 |
+
"eval_loss": 1.6813957691192627,
|
| 870 |
+
"eval_runtime": 53.307,
|
| 871 |
+
"eval_samples_per_second": 9.079,
|
| 872 |
+
"eval_steps_per_second": 2.27,
|
| 873 |
+
"step": 600
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.97,
|
| 877 |
+
"grad_norm": 0.6295060515403748,
|
| 878 |
+
"learning_rate": 0.0002589059831878262,
|
| 879 |
+
"loss": 1.6353,
|
| 880 |
+
"step": 605
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 0.97,
|
| 884 |
+
"grad_norm": 0.6981686353683472,
|
| 885 |
+
"learning_rate": 0.00025793564557730065,
|
| 886 |
+
"loss": 1.6768,
|
| 887 |
+
"step": 610
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 0.98,
|
| 891 |
+
"grad_norm": 0.6336885094642639,
|
| 892 |
+
"learning_rate": 0.0002569558495692474,
|
| 893 |
+
"loss": 1.6594,
|
| 894 |
+
"step": 615
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 0.99,
|
| 898 |
+
"grad_norm": 0.6425840258598328,
|
| 899 |
+
"learning_rate": 0.00025596668102316266,
|
| 900 |
+
"loss": 1.7709,
|
| 901 |
+
"step": 620
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 1.0,
|
| 905 |
+
"grad_norm": 0.5762469172477722,
|
| 906 |
+
"learning_rate": 0.0002549682266198577,
|
| 907 |
+
"loss": 1.599,
|
| 908 |
+
"step": 625
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 1.01,
|
| 912 |
+
"grad_norm": 0.6281951069831848,
|
| 913 |
+
"learning_rate": 0.00025396057385386326,
|
| 914 |
+
"loss": 1.6798,
|
| 915 |
+
"step": 630
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 1.01,
|
| 919 |
+
"grad_norm": 0.6240226030349731,
|
| 920 |
+
"learning_rate": 0.0002529438110257623,
|
| 921 |
+
"loss": 1.491,
|
| 922 |
+
"step": 635
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 1.02,
|
| 926 |
+
"grad_norm": 0.7243877053260803,
|
| 927 |
+
"learning_rate": 0.00025191802723445233,
|
| 928 |
+
"loss": 1.6763,
|
| 929 |
+
"step": 640
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 1.03,
|
| 933 |
+
"grad_norm": 0.7280773520469666,
|
| 934 |
+
"learning_rate": 0.0002508833123693376,
|
| 935 |
+
"loss": 1.5093,
|
| 936 |
+
"step": 645
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 1.04,
|
| 940 |
+
"grad_norm": 0.6628164649009705,
|
| 941 |
+
"learning_rate": 0.0002498397571024521,
|
| 942 |
+
"loss": 1.5736,
|
| 943 |
+
"step": 650
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 1.05,
|
| 947 |
+
"grad_norm": 0.8163209557533264,
|
| 948 |
+
"learning_rate": 0.00024878745288051384,
|
| 949 |
+
"loss": 1.6704,
|
| 950 |
+
"step": 655
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 1.05,
|
| 954 |
+
"grad_norm": 0.7377064228057861,
|
| 955 |
+
"learning_rate": 0.0002477264919169117,
|
| 956 |
+
"loss": 1.6146,
|
| 957 |
+
"step": 660
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 1.06,
|
| 961 |
+
"grad_norm": 0.7334110140800476,
|
| 962 |
+
"learning_rate": 0.0002466569671836246,
|
| 963 |
+
"loss": 1.5641,
|
| 964 |
+
"step": 665
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 1.07,
|
| 968 |
+
"grad_norm": 0.708491325378418,
|
| 969 |
+
"learning_rate": 0.000245578972403074,
|
| 970 |
+
"loss": 1.6447,
|
| 971 |
+
"step": 670
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 1.08,
|
| 975 |
+
"grad_norm": 0.688526451587677,
|
| 976 |
+
"learning_rate": 0.0002444926020399118,
|
| 977 |
+
"loss": 1.4617,
|
| 978 |
+
"step": 675
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 1.08,
|
| 982 |
+
"grad_norm": 0.6798132658004761,
|
| 983 |
+
"learning_rate": 0.00024339795129274178,
|
| 984 |
+
"loss": 1.4603,
|
| 985 |
+
"step": 680
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 1.09,
|
| 989 |
+
"grad_norm": 0.7954903244972229,
|
| 990 |
+
"learning_rate": 0.0002422951160857775,
|
| 991 |
+
"loss": 1.6878,
|
| 992 |
+
"step": 685
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 1.1,
|
| 996 |
+
"grad_norm": 0.7106612920761108,
|
| 997 |
+
"learning_rate": 0.00024118419306043653,
|
| 998 |
+
"loss": 1.5761,
|
| 999 |
+
"step": 690
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 1.11,
|
| 1003 |
+
"grad_norm": 0.6339300870895386,
|
| 1004 |
+
"learning_rate": 0.00024006527956687171,
|
| 1005 |
+
"loss": 1.6388,
|
| 1006 |
+
"step": 695
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 1.12,
|
| 1010 |
+
"grad_norm": 0.6403482556343079,
|
| 1011 |
+
"learning_rate": 0.00023893847365544043,
|
| 1012 |
+
"loss": 1.6328,
|
| 1013 |
+
"step": 700
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 1.12,
|
| 1017 |
+
"grad_norm": 0.6963227391242981,
|
| 1018 |
+
"learning_rate": 0.0002378038740681124,
|
| 1019 |
+
"loss": 1.5307,
|
| 1020 |
+
"step": 705
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 1.13,
|
| 1024 |
+
"grad_norm": 0.7481818795204163,
|
| 1025 |
+
"learning_rate": 0.00023666158022981696,
|
| 1026 |
+
"loss": 1.544,
|
| 1027 |
+
"step": 710
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 1.14,
|
| 1031 |
+
"grad_norm": 0.6810987591743469,
|
| 1032 |
+
"learning_rate": 0.00023551169223973037,
|
| 1033 |
+
"loss": 1.4045,
|
| 1034 |
+
"step": 715
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 1.15,
|
| 1038 |
+
"grad_norm": 0.7195698022842407,
|
| 1039 |
+
"learning_rate": 0.00023435431086250424,
|
| 1040 |
+
"loss": 1.5408,
|
| 1041 |
+
"step": 720
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 1.16,
|
| 1045 |
+
"grad_norm": 0.8156709671020508,
|
| 1046 |
+
"learning_rate": 0.0002331895375194355,
|
| 1047 |
+
"loss": 1.5569,
|
| 1048 |
+
"step": 725
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 1.16,
|
| 1052 |
+
"grad_norm": 0.6353544592857361,
|
| 1053 |
+
"learning_rate": 0.00023201747427957882,
|
| 1054 |
+
"loss": 1.5852,
|
| 1055 |
+
"step": 730
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 1.17,
|
| 1059 |
+
"grad_norm": 0.7116926312446594,
|
| 1060 |
+
"learning_rate": 0.0002308382238508023,
|
| 1061 |
+
"loss": 1.5359,
|
| 1062 |
+
"step": 735
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 1.18,
|
| 1066 |
+
"grad_norm": 0.7053561806678772,
|
| 1067 |
+
"learning_rate": 0.0002296518895707871,
|
| 1068 |
+
"loss": 1.6377,
|
| 1069 |
+
"step": 740
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 1.19,
|
| 1073 |
+
"grad_norm": 0.8180266618728638,
|
| 1074 |
+
"learning_rate": 0.0002284585753979723,
|
| 1075 |
+
"loss": 1.5982,
|
| 1076 |
+
"step": 745
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 1.2,
|
| 1080 |
+
"grad_norm": 0.6952159404754639,
|
| 1081 |
+
"learning_rate": 0.00022725838590244472,
|
| 1082 |
+
"loss": 1.5485,
|
| 1083 |
+
"step": 750
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 1.2,
|
| 1087 |
+
"grad_norm": 0.8135376572608948,
|
| 1088 |
+
"learning_rate": 0.00022605142625677525,
|
| 1089 |
+
"loss": 1.6058,
|
| 1090 |
+
"step": 755
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 1.21,
|
| 1094 |
+
"grad_norm": 0.6240015625953674,
|
| 1095 |
+
"learning_rate": 0.00022483780222680314,
|
| 1096 |
+
"loss": 1.6065,
|
| 1097 |
+
"step": 760
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 1.22,
|
| 1101 |
+
"grad_norm": 0.8067346215248108,
|
| 1102 |
+
"learning_rate": 0.00022361762016236738,
|
| 1103 |
+
"loss": 1.5782,
|
| 1104 |
+
"step": 765
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 1.23,
|
| 1108 |
+
"grad_norm": 0.7767935991287231,
|
| 1109 |
+
"learning_rate": 0.0002223909869879872,
|
| 1110 |
+
"loss": 1.4984,
|
| 1111 |
+
"step": 770
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 1.24,
|
| 1115 |
+
"grad_norm": 0.7751308083534241,
|
| 1116 |
+
"learning_rate": 0.00022115801019349256,
|
| 1117 |
+
"loss": 1.6065,
|
| 1118 |
+
"step": 775
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 1.24,
|
| 1122 |
+
"grad_norm": 0.6872211694717407,
|
| 1123 |
+
"learning_rate": 0.00021991879782460468,
|
| 1124 |
+
"loss": 1.6571,
|
| 1125 |
+
"step": 780
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 1.25,
|
| 1129 |
+
"grad_norm": 0.6883332133293152,
|
| 1130 |
+
"learning_rate": 0.00021867345847346783,
|
| 1131 |
+
"loss": 1.587,
|
| 1132 |
+
"step": 785
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 1.26,
|
| 1136 |
+
"grad_norm": 0.6802603006362915,
|
| 1137 |
+
"learning_rate": 0.00021742210126913364,
|
| 1138 |
+
"loss": 1.527,
|
| 1139 |
+
"step": 790
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 1.27,
|
| 1143 |
+
"grad_norm": 0.8581397533416748,
|
| 1144 |
+
"learning_rate": 0.0002161648358679981,
|
| 1145 |
+
"loss": 1.5975,
|
| 1146 |
+
"step": 795
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 1.28,
|
| 1150 |
+
"grad_norm": 0.779639482498169,
|
| 1151 |
+
"learning_rate": 0.00021490177244419225,
|
| 1152 |
+
"loss": 1.5361,
|
| 1153 |
+
"step": 800
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 1.28,
|
| 1157 |
+
"eval_loss": 1.6755969524383545,
|
| 1158 |
+
"eval_runtime": 54.1533,
|
| 1159 |
+
"eval_samples_per_second": 8.938,
|
| 1160 |
+
"eval_steps_per_second": 2.234,
|
| 1161 |
+
"step": 800
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 1.28,
|
| 1165 |
+
"grad_norm": 0.7144075036048889,
|
| 1166 |
+
"learning_rate": 0.00021363302167992773,
|
| 1167 |
+
"loss": 1.6061,
|
| 1168 |
+
"step": 805
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 1.29,
|
| 1172 |
+
"grad_norm": 0.7203413844108582,
|
| 1173 |
+
"learning_rate": 0.00021235869475579763,
|
| 1174 |
+
"loss": 1.5182,
|
| 1175 |
+
"step": 810
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 1.3,
|
| 1179 |
+
"grad_norm": 0.7165353298187256,
|
| 1180 |
+
"learning_rate": 0.00021107890334103372,
|
| 1181 |
+
"loss": 1.4595,
|
| 1182 |
+
"step": 815
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 1.31,
|
| 1186 |
+
"grad_norm": 0.7226712703704834,
|
| 1187 |
+
"learning_rate": 0.0002097937595837211,
|
| 1188 |
+
"loss": 1.5236,
|
| 1189 |
+
"step": 820
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 1.32,
|
| 1193 |
+
"grad_norm": 0.6462448239326477,
|
| 1194 |
+
"learning_rate": 0.00020850337610097027,
|
| 1195 |
+
"loss": 1.6672,
|
| 1196 |
+
"step": 825
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 1.32,
|
| 1200 |
+
"grad_norm": 0.8719556331634521,
|
| 1201 |
+
"learning_rate": 0.00020720786596904893,
|
| 1202 |
+
"loss": 1.5598,
|
| 1203 |
+
"step": 830
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 1.33,
|
| 1207 |
+
"grad_norm": 0.8658766150474548,
|
| 1208 |
+
"learning_rate": 0.0002059073427134727,
|
| 1209 |
+
"loss": 1.6314,
|
| 1210 |
+
"step": 835
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 1.34,
|
| 1214 |
+
"grad_norm": 0.6820198893547058,
|
| 1215 |
+
"learning_rate": 0.00020460192029905726,
|
| 1216 |
+
"loss": 1.5311,
|
| 1217 |
+
"step": 840
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 1.35,
|
| 1221 |
+
"grad_norm": 0.6817758083343506,
|
| 1222 |
+
"learning_rate": 0.00020329171311993135,
|
| 1223 |
+
"loss": 1.4233,
|
| 1224 |
+
"step": 845
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 1.36,
|
| 1228 |
+
"grad_norm": 0.7133145928382874,
|
| 1229 |
+
"learning_rate": 0.00020197683598951248,
|
| 1230 |
+
"loss": 1.5895,
|
| 1231 |
+
"step": 850
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 1.36,
|
| 1235 |
+
"grad_norm": 0.684045135974884,
|
| 1236 |
+
"learning_rate": 0.00020065740413044588,
|
| 1237 |
+
"loss": 1.7072,
|
| 1238 |
+
"step": 855
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 1.37,
|
| 1242 |
+
"grad_norm": 0.8078965544700623,
|
| 1243 |
+
"learning_rate": 0.0001993335331645076,
|
| 1244 |
+
"loss": 1.6305,
|
| 1245 |
+
"step": 860
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 1.38,
|
| 1249 |
+
"grad_norm": 0.7130747437477112,
|
| 1250 |
+
"learning_rate": 0.00019800533910247243,
|
| 1251 |
+
"loss": 1.5375,
|
| 1252 |
+
"step": 865
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 1.39,
|
| 1256 |
+
"grad_norm": 0.8379193544387817,
|
| 1257 |
+
"learning_rate": 0.00019667293833394793,
|
| 1258 |
+
"loss": 1.5368,
|
| 1259 |
+
"step": 870
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 1.4,
|
| 1263 |
+
"grad_norm": 0.8084167242050171,
|
| 1264 |
+
"learning_rate": 0.00019533644761717518,
|
| 1265 |
+
"loss": 1.729,
|
| 1266 |
+
"step": 875
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 1.4,
|
| 1270 |
+
"grad_norm": 0.8069108724594116,
|
| 1271 |
+
"learning_rate": 0.00019399598406879748,
|
| 1272 |
+
"loss": 1.5978,
|
| 1273 |
+
"step": 880
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 1.41,
|
| 1277 |
+
"grad_norm": 0.8795610666275024,
|
| 1278 |
+
"learning_rate": 0.0001926516651535971,
|
| 1279 |
+
"loss": 1.7128,
|
| 1280 |
+
"step": 885
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 1.42,
|
| 1284 |
+
"grad_norm": 0.7351583242416382,
|
| 1285 |
+
"learning_rate": 0.00019130360867420203,
|
| 1286 |
+
"loss": 1.6878,
|
| 1287 |
+
"step": 890
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 1.43,
|
| 1291 |
+
"grad_norm": 0.747438907623291,
|
| 1292 |
+
"learning_rate": 0.00018995193276076294,
|
| 1293 |
+
"loss": 1.4505,
|
| 1294 |
+
"step": 895
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 1.44,
|
| 1298 |
+
"grad_norm": 0.7769878506660461,
|
| 1299 |
+
"learning_rate": 0.00018859675586060134,
|
| 1300 |
+
"loss": 1.4913,
|
| 1301 |
+
"step": 900
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 1.44,
|
| 1305 |
+
"grad_norm": 0.7609659433364868,
|
| 1306 |
+
"learning_rate": 0.0001872381967278302,
|
| 1307 |
+
"loss": 1.6376,
|
| 1308 |
+
"step": 905
|
| 1309 |
+
},
|
| 1310 |
+
{
|
| 1311 |
+
"epoch": 1.45,
|
| 1312 |
+
"grad_norm": 0.7220240831375122,
|
| 1313 |
+
"learning_rate": 0.00018587637441294734,
|
| 1314 |
+
"loss": 1.5796,
|
| 1315 |
+
"step": 910
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 1.46,
|
| 1319 |
+
"grad_norm": 0.7993665337562561,
|
| 1320 |
+
"learning_rate": 0.00018451140825240312,
|
| 1321 |
+
"loss": 1.5595,
|
| 1322 |
+
"step": 915
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 1.47,
|
| 1326 |
+
"grad_norm": 0.7627410292625427,
|
| 1327 |
+
"learning_rate": 0.00018314341785814317,
|
| 1328 |
+
"loss": 1.618,
|
| 1329 |
+
"step": 920
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 1.48,
|
| 1333 |
+
"grad_norm": 0.7644938826560974,
|
| 1334 |
+
"learning_rate": 0.00018177252310712646,
|
| 1335 |
+
"loss": 1.5943,
|
| 1336 |
+
"step": 925
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 1.48,
|
| 1340 |
+
"grad_norm": 0.6797645688056946,
|
| 1341 |
+
"learning_rate": 0.00018039884413082065,
|
| 1342 |
+
"loss": 1.5168,
|
| 1343 |
+
"step": 930
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 1.49,
|
| 1347 |
+
"grad_norm": 0.7855983376502991,
|
| 1348 |
+
"learning_rate": 0.00017902250130467524,
|
| 1349 |
+
"loss": 1.3884,
|
| 1350 |
+
"step": 935
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 1.5,
|
| 1354 |
+
"grad_norm": 0.7340678572654724,
|
| 1355 |
+
"learning_rate": 0.00017764361523757246,
|
| 1356 |
+
"loss": 1.6262,
|
| 1357 |
+
"step": 940
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 1.51,
|
| 1361 |
+
"grad_norm": 0.9311641454696655,
|
| 1362 |
+
"learning_rate": 0.00017626230676125887,
|
| 1363 |
+
"loss": 1.5546,
|
| 1364 |
+
"step": 945
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 1.52,
|
| 1368 |
+
"grad_norm": 0.7892611622810364,
|
| 1369 |
+
"learning_rate": 0.0001748786969197567,
|
| 1370 |
+
"loss": 1.541,
|
| 1371 |
+
"step": 950
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 1.52,
|
| 1375 |
+
"grad_norm": 0.737107515335083,
|
| 1376 |
+
"learning_rate": 0.00017349290695875668,
|
| 1377 |
+
"loss": 1.543,
|
| 1378 |
+
"step": 955
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 1.53,
|
| 1382 |
+
"grad_norm": 0.7268142104148865,
|
| 1383 |
+
"learning_rate": 0.00017210505831499336,
|
| 1384 |
+
"loss": 1.3838,
|
| 1385 |
+
"step": 960
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 1.54,
|
| 1389 |
+
"grad_norm": 0.7059959769248962,
|
| 1390 |
+
"learning_rate": 0.00017071527260560356,
|
| 1391 |
+
"loss": 1.4105,
|
| 1392 |
+
"step": 965
|
| 1393 |
+
},
|
| 1394 |
+
{
|
| 1395 |
+
"epoch": 1.55,
|
| 1396 |
+
"grad_norm": 0.7432154417037964,
|
| 1397 |
+
"learning_rate": 0.00016932367161746922,
|
| 1398 |
+
"loss": 1.5525,
|
| 1399 |
+
"step": 970
|
| 1400 |
+
},
|
| 1401 |
+
{
|
| 1402 |
+
"epoch": 1.56,
|
| 1403 |
+
"grad_norm": 0.7598636746406555,
|
| 1404 |
+
"learning_rate": 0.00016793037729654503,
|
| 1405 |
+
"loss": 1.4841,
|
| 1406 |
+
"step": 975
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 1.56,
|
| 1410 |
+
"grad_norm": 0.7130389213562012,
|
| 1411 |
+
"learning_rate": 0.0001665355117371723,
|
| 1412 |
+
"loss": 1.6633,
|
| 1413 |
+
"step": 980
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 1.57,
|
| 1417 |
+
"grad_norm": 0.7120307087898254,
|
| 1418 |
+
"learning_rate": 0.00016513919717138015,
|
| 1419 |
+
"loss": 1.661,
|
| 1420 |
+
"step": 985
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 1.58,
|
| 1424 |
+
"grad_norm": 0.7838018536567688,
|
| 1425 |
+
"learning_rate": 0.00016374155595817383,
|
| 1426 |
+
"loss": 1.4154,
|
| 1427 |
+
"step": 990
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 1.59,
|
| 1431 |
+
"grad_norm": 0.8129784464836121,
|
| 1432 |
+
"learning_rate": 0.00016234271057281288,
|
| 1433 |
+
"loss": 1.4759,
|
| 1434 |
+
"step": 995
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 1.6,
|
| 1438 |
+
"grad_norm": 0.7084518671035767,
|
| 1439 |
+
"learning_rate": 0.0001609427835960783,
|
| 1440 |
+
"loss": 1.5625,
|
| 1441 |
+
"step": 1000
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 1.6,
|
| 1445 |
+
"eval_loss": 1.656510591506958,
|
| 1446 |
+
"eval_runtime": 53.8113,
|
| 1447 |
+
"eval_samples_per_second": 8.994,
|
| 1448 |
+
"eval_steps_per_second": 2.249,
|
| 1449 |
+
"step": 1000
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 1.6,
|
| 1453 |
+
"grad_norm": 0.7194417715072632,
|
| 1454 |
+
"learning_rate": 0.00015954189770353086,
|
| 1455 |
+
"loss": 1.6973,
|
| 1456 |
+
"step": 1005
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 1.61,
|
| 1460 |
+
"grad_norm": 0.8406971096992493,
|
| 1461 |
+
"learning_rate": 0.00015814017565476122,
|
| 1462 |
+
"loss": 1.5086,
|
| 1463 |
+
"step": 1010
|
| 1464 |
+
},
|
| 1465 |
+
{
|
| 1466 |
+
"epoch": 1.62,
|
| 1467 |
+
"grad_norm": 0.7156040072441101,
|
| 1468 |
+
"learning_rate": 0.0001567377402826323,
|
| 1469 |
+
"loss": 1.6073,
|
| 1470 |
+
"step": 1015
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 1.63,
|
| 1474 |
+
"grad_norm": 0.7198348045349121,
|
| 1475 |
+
"learning_rate": 0.00015533471448251555,
|
| 1476 |
+
"loss": 1.5575,
|
| 1477 |
+
"step": 1020
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 1.64,
|
| 1481 |
+
"grad_norm": 0.7150219678878784,
|
| 1482 |
+
"learning_rate": 0.0001539312212015217,
|
| 1483 |
+
"loss": 1.5192,
|
| 1484 |
+
"step": 1025
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 1.64,
|
| 1488 |
+
"grad_norm": 0.6892033815383911,
|
| 1489 |
+
"learning_rate": 0.0001525273834277266,
|
| 1490 |
+
"loss": 1.4947,
|
| 1491 |
+
"step": 1030
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 1.65,
|
| 1495 |
+
"grad_norm": 0.7327867746353149,
|
| 1496 |
+
"learning_rate": 0.00015112332417939418,
|
| 1497 |
+
"loss": 1.5601,
|
| 1498 |
+
"step": 1035
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 1.66,
|
| 1502 |
+
"grad_norm": 0.6658421754837036,
|
| 1503 |
+
"learning_rate": 0.00014971916649419615,
|
| 1504 |
+
"loss": 1.6199,
|
| 1505 |
+
"step": 1040
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 1.67,
|
| 1509 |
+
"grad_norm": 0.8288849592208862,
|
| 1510 |
+
"learning_rate": 0.00014831503341843018,
|
| 1511 |
+
"loss": 1.6261,
|
| 1512 |
+
"step": 1045
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 1.68,
|
| 1516 |
+
"grad_norm": 0.7870001196861267,
|
| 1517 |
+
"learning_rate": 0.00014691104799623744,
|
| 1518 |
+
"loss": 1.5233,
|
| 1519 |
+
"step": 1050
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 1.68,
|
| 1523 |
+
"grad_norm": 0.7046535015106201,
|
| 1524 |
+
"learning_rate": 0.00014550733325882024,
|
| 1525 |
+
"loss": 1.6226,
|
| 1526 |
+
"step": 1055
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 1.69,
|
| 1530 |
+
"grad_norm": 0.7216707468032837,
|
| 1531 |
+
"learning_rate": 0.0001441040122136608,
|
| 1532 |
+
"loss": 1.4483,
|
| 1533 |
+
"step": 1060
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 1.7,
|
| 1537 |
+
"grad_norm": 0.7812987565994263,
|
| 1538 |
+
"learning_rate": 0.00014270120783374204,
|
| 1539 |
+
"loss": 1.512,
|
| 1540 |
+
"step": 1065
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 1.71,
|
| 1544 |
+
"grad_norm": 0.7140246629714966,
|
| 1545 |
+
"learning_rate": 0.0001412990430467715,
|
| 1546 |
+
"loss": 1.5703,
|
| 1547 |
+
"step": 1070
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 1.72,
|
| 1551 |
+
"grad_norm": 0.8250936269760132,
|
| 1552 |
+
"learning_rate": 0.00013989764072440945,
|
| 1553 |
+
"loss": 1.5059,
|
| 1554 |
+
"step": 1075
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 1.72,
|
| 1558 |
+
"grad_norm": 0.7511314749717712,
|
| 1559 |
+
"learning_rate": 0.000138497123671501,
|
| 1560 |
+
"loss": 1.6571,
|
| 1561 |
+
"step": 1080
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 1.73,
|
| 1565 |
+
"grad_norm": 0.7608028054237366,
|
| 1566 |
+
"learning_rate": 0.00013709761461531552,
|
| 1567 |
+
"loss": 1.4981,
|
| 1568 |
+
"step": 1085
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 1.74,
|
| 1572 |
+
"grad_norm": 0.8892323970794678,
|
| 1573 |
+
"learning_rate": 0.00013569923619479138,
|
| 1574 |
+
"loss": 1.5202,
|
| 1575 |
+
"step": 1090
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 1.75,
|
| 1579 |
+
"grad_norm": 0.7229086756706238,
|
| 1580 |
+
"learning_rate": 0.00013430211094978955,
|
| 1581 |
+
"loss": 1.5342,
|
| 1582 |
+
"step": 1095
|
| 1583 |
+
},
|
| 1584 |
+
{
|
| 1585 |
+
"epoch": 1.76,
|
| 1586 |
+
"grad_norm": 0.814064621925354,
|
| 1587 |
+
"learning_rate": 0.00013290636131035513,
|
| 1588 |
+
"loss": 1.4798,
|
| 1589 |
+
"step": 1100
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 1.76,
|
| 1593 |
+
"grad_norm": 0.5819219350814819,
|
| 1594 |
+
"learning_rate": 0.0001315121095859891,
|
| 1595 |
+
"loss": 1.3957,
|
| 1596 |
+
"step": 1105
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 1.77,
|
| 1600 |
+
"grad_norm": 0.7352663278579712,
|
| 1601 |
+
"learning_rate": 0.00013011947795492994,
|
| 1602 |
+
"loss": 1.5094,
|
| 1603 |
+
"step": 1110
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 1.78,
|
| 1607 |
+
"grad_norm": 0.7600275874137878,
|
| 1608 |
+
"learning_rate": 0.0001287285884534478,
|
| 1609 |
+
"loss": 1.4571,
|
| 1610 |
+
"step": 1115
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 1.79,
|
| 1614 |
+
"grad_norm": 0.795772910118103,
|
| 1615 |
+
"learning_rate": 0.00012733956296514954,
|
| 1616 |
+
"loss": 1.5675,
|
| 1617 |
+
"step": 1120
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 1.79,
|
| 1621 |
+
"grad_norm": 0.8060831427574158,
|
| 1622 |
+
"learning_rate": 0.0001259525232102991,
|
| 1623 |
+
"loss": 1.4759,
|
| 1624 |
+
"step": 1125
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 1.8,
|
| 1628 |
+
"grad_norm": 0.8213143348693848,
|
| 1629 |
+
"learning_rate": 0.0001245675907351503,
|
| 1630 |
+
"loss": 1.6123,
|
| 1631 |
+
"step": 1130
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 1.81,
|
| 1635 |
+
"grad_norm": 0.7344096302986145,
|
| 1636 |
+
"learning_rate": 0.00012318488690129633,
|
| 1637 |
+
"loss": 1.5137,
|
| 1638 |
+
"step": 1135
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 1.82,
|
| 1642 |
+
"grad_norm": 0.6787902116775513,
|
| 1643 |
+
"learning_rate": 0.00012180453287503444,
|
| 1644 |
+
"loss": 1.6397,
|
| 1645 |
+
"step": 1140
|
| 1646 |
+
},
|
| 1647 |
+
{
|
| 1648 |
+
"epoch": 1.83,
|
| 1649 |
+
"grad_norm": 0.7361781001091003,
|
| 1650 |
+
"learning_rate": 0.00012042664961674843,
|
| 1651 |
+
"loss": 1.453,
|
| 1652 |
+
"step": 1145
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 1.83,
|
| 1656 |
+
"grad_norm": 0.7698022127151489,
|
| 1657 |
+
"learning_rate": 0.00011905135787030865,
|
| 1658 |
+
"loss": 1.4961,
|
| 1659 |
+
"step": 1150
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 1.84,
|
| 1663 |
+
"grad_norm": 0.7049584984779358,
|
| 1664 |
+
"learning_rate": 0.00011767877815249155,
|
| 1665 |
+
"loss": 1.5562,
|
| 1666 |
+
"step": 1155
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 1.85,
|
| 1670 |
+
"grad_norm": 0.7255071401596069,
|
| 1671 |
+
"learning_rate": 0.00011630903074241826,
|
| 1672 |
+
"loss": 1.6307,
|
| 1673 |
+
"step": 1160
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 1.86,
|
| 1677 |
+
"grad_norm": 0.7467817664146423,
|
| 1678 |
+
"learning_rate": 0.0001149422356710152,
|
| 1679 |
+
"loss": 1.4051,
|
| 1680 |
+
"step": 1165
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 1.87,
|
| 1684 |
+
"grad_norm": 0.6346266269683838,
|
| 1685 |
+
"learning_rate": 0.00011357851271049528,
|
| 1686 |
+
"loss": 1.6404,
|
| 1687 |
+
"step": 1170
|
| 1688 |
+
},
|
| 1689 |
+
{
|
| 1690 |
+
"epoch": 1.87,
|
| 1691 |
+
"grad_norm": 0.6892123818397522,
|
| 1692 |
+
"learning_rate": 0.00011221798136386238,
|
| 1693 |
+
"loss": 1.462,
|
| 1694 |
+
"step": 1175
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 1.88,
|
| 1698 |
+
"grad_norm": 0.7803078889846802,
|
| 1699 |
+
"learning_rate": 0.00011086076085443949,
|
| 1700 |
+
"loss": 1.6246,
|
| 1701 |
+
"step": 1180
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 1.89,
|
| 1705 |
+
"grad_norm": 0.6938590407371521,
|
| 1706 |
+
"learning_rate": 0.00010950697011542082,
|
| 1707 |
+
"loss": 1.3757,
|
| 1708 |
+
"step": 1185
|
| 1709 |
+
},
|
| 1710 |
+
{
|
| 1711 |
+
"epoch": 1.9,
|
| 1712 |
+
"grad_norm": 0.6958265900611877,
|
| 1713 |
+
"learning_rate": 0.00010815672777945013,
|
| 1714 |
+
"loss": 1.3996,
|
| 1715 |
+
"step": 1190
|
| 1716 |
+
},
|
| 1717 |
+
{
|
| 1718 |
+
"epoch": 1.91,
|
| 1719 |
+
"grad_norm": 0.9197216033935547,
|
| 1720 |
+
"learning_rate": 0.00010681015216822454,
|
| 1721 |
+
"loss": 1.5832,
|
| 1722 |
+
"step": 1195
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 1.91,
|
| 1726 |
+
"grad_norm": 0.6959114670753479,
|
| 1727 |
+
"learning_rate": 0.00010546736128212632,
|
| 1728 |
+
"loss": 1.606,
|
| 1729 |
+
"step": 1200
|
| 1730 |
+
},
|
| 1731 |
+
{
|
| 1732 |
+
"epoch": 1.91,
|
| 1733 |
+
"eval_loss": 1.6396043300628662,
|
| 1734 |
+
"eval_runtime": 53.5893,
|
| 1735 |
+
"eval_samples_per_second": 9.032,
|
| 1736 |
+
"eval_steps_per_second": 2.258,
|
| 1737 |
+
"step": 1200
|
| 1738 |
+
},
|
| 1739 |
+
{
|
| 1740 |
+
"epoch": 1.92,
|
| 1741 |
+
"grad_norm": 0.8191283345222473,
|
| 1742 |
+
"learning_rate": 0.00010412847278988228,
|
| 1743 |
+
"loss": 1.5161,
|
| 1744 |
+
"step": 1205
|
| 1745 |
+
},
|
| 1746 |
+
{
|
| 1747 |
+
"epoch": 1.93,
|
| 1748 |
+
"grad_norm": 0.9132874011993408,
|
| 1749 |
+
"learning_rate": 0.00010279360401825263,
|
| 1750 |
+
"loss": 1.4927,
|
| 1751 |
+
"step": 1210
|
| 1752 |
+
},
|
| 1753 |
+
{
|
| 1754 |
+
"epoch": 1.94,
|
| 1755 |
+
"grad_norm": 0.7479792833328247,
|
| 1756 |
+
"learning_rate": 0.00010146287194174953,
|
| 1757 |
+
"loss": 1.5343,
|
| 1758 |
+
"step": 1215
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 1.95,
|
| 1762 |
+
"grad_norm": 0.7714102864265442,
|
| 1763 |
+
"learning_rate": 0.00010013639317238674,
|
| 1764 |
+
"loss": 1.6156,
|
| 1765 |
+
"step": 1220
|
| 1766 |
+
},
|
| 1767 |
+
{
|
| 1768 |
+
"epoch": 1.95,
|
| 1769 |
+
"grad_norm": 0.8507571220397949,
|
| 1770 |
+
"learning_rate": 9.88142839494607e-05,
|
| 1771 |
+
"loss": 1.6551,
|
| 1772 |
+
"step": 1225
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 1.96,
|
| 1776 |
+
"grad_norm": 0.8553720116615295,
|
| 1777 |
+
"learning_rate": 9.749666012936487e-05,
|
| 1778 |
+
"loss": 1.6274,
|
| 1779 |
+
"step": 1230
|
| 1780 |
+
},
|
| 1781 |
+
{
|
| 1782 |
+
"epoch": 1.97,
|
| 1783 |
+
"grad_norm": 0.6895230412483215,
|
| 1784 |
+
"learning_rate": 9.618363717543669e-05,
|
| 1785 |
+
"loss": 1.5211,
|
| 1786 |
+
"step": 1235
|
| 1787 |
+
},
|
| 1788 |
+
{
|
| 1789 |
+
"epoch": 1.98,
|
| 1790 |
+
"grad_norm": 0.6788089871406555,
|
| 1791 |
+
"learning_rate": 9.487533014784007e-05,
|
| 1792 |
+
"loss": 1.4057,
|
| 1793 |
+
"step": 1240
|
| 1794 |
+
},
|
| 1795 |
+
{
|
| 1796 |
+
"epoch": 1.99,
|
| 1797 |
+
"grad_norm": 0.851324737071991,
|
| 1798 |
+
"learning_rate": 9.357185369348225e-05,
|
| 1799 |
+
"loss": 1.3783,
|
| 1800 |
+
"step": 1245
|
| 1801 |
+
},
|
| 1802 |
+
{
|
| 1803 |
+
"epoch": 1.99,
|
| 1804 |
+
"grad_norm": 0.7894309163093567,
|
| 1805 |
+
"learning_rate": 9.227332203596764e-05,
|
| 1806 |
+
"loss": 1.5338,
|
| 1807 |
+
"step": 1250
|
| 1808 |
+
},
|
| 1809 |
+
{
|
| 1810 |
+
"epoch": 2.0,
|
| 1811 |
+
"grad_norm": 0.6584810018539429,
|
| 1812 |
+
"learning_rate": 9.097984896558807e-05,
|
| 1813 |
+
"loss": 1.4221,
|
| 1814 |
+
"step": 1255
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 2.01,
|
| 1818 |
+
"grad_norm": 0.6491633653640747,
|
| 1819 |
+
"learning_rate": 8.96915478293517e-05,
|
| 1820 |
+
"loss": 1.4814,
|
| 1821 |
+
"step": 1260
|
| 1822 |
+
},
|
| 1823 |
+
{
|
| 1824 |
+
"epoch": 2.02,
|
| 1825 |
+
"grad_norm": 0.8882700800895691,
|
| 1826 |
+
"learning_rate": 8.840853152105004e-05,
|
| 1827 |
+
"loss": 1.4217,
|
| 1828 |
+
"step": 1265
|
| 1829 |
+
},
|
| 1830 |
+
{
|
| 1831 |
+
"epoch": 2.03,
|
| 1832 |
+
"grad_norm": 0.8957266807556152,
|
| 1833 |
+
"learning_rate": 8.713091247136558e-05,
|
| 1834 |
+
"loss": 1.4734,
|
| 1835 |
+
"step": 1270
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 2.03,
|
| 1839 |
+
"grad_norm": 0.8561681509017944,
|
| 1840 |
+
"learning_rate": 8.585880263801893e-05,
|
| 1841 |
+
"loss": 1.4981,
|
| 1842 |
+
"step": 1275
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 2.04,
|
| 1846 |
+
"grad_norm": 0.8122314810752869,
|
| 1847 |
+
"learning_rate": 8.459231349595824e-05,
|
| 1848 |
+
"loss": 1.3924,
|
| 1849 |
+
"step": 1280
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 2.05,
|
| 1853 |
+
"grad_norm": 0.7481365203857422,
|
| 1854 |
+
"learning_rate": 8.333155602759088e-05,
|
| 1855 |
+
"loss": 1.5153,
|
| 1856 |
+
"step": 1285
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 2.06,
|
| 1860 |
+
"grad_norm": 0.9371137619018555,
|
| 1861 |
+
"learning_rate": 8.207664071305751e-05,
|
| 1862 |
+
"loss": 1.4867,
|
| 1863 |
+
"step": 1290
|
| 1864 |
+
},
|
| 1865 |
+
{
|
| 1866 |
+
"epoch": 2.07,
|
| 1867 |
+
"grad_norm": 0.8563275337219238,
|
| 1868 |
+
"learning_rate": 8.082767752055129e-05,
|
| 1869 |
+
"loss": 1.4395,
|
| 1870 |
+
"step": 1295
|
| 1871 |
+
},
|
| 1872 |
+
{
|
| 1873 |
+
"epoch": 2.07,
|
| 1874 |
+
"grad_norm": 0.7887305617332458,
|
| 1875 |
+
"learning_rate": 7.958477589668092e-05,
|
| 1876 |
+
"loss": 1.4446,
|
| 1877 |
+
"step": 1300
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 2.08,
|
| 1881 |
+
"grad_norm": 0.852190375328064,
|
| 1882 |
+
"learning_rate": 7.834804475688008e-05,
|
| 1883 |
+
"loss": 1.3596,
|
| 1884 |
+
"step": 1305
|
| 1885 |
+
},
|
| 1886 |
+
{
|
| 1887 |
+
"epoch": 2.09,
|
| 1888 |
+
"grad_norm": 0.7616653442382812,
|
| 1889 |
+
"learning_rate": 7.711759247586301e-05,
|
| 1890 |
+
"loss": 1.452,
|
| 1891 |
+
"step": 1310
|
| 1892 |
+
},
|
| 1893 |
+
{
|
| 1894 |
+
"epoch": 2.1,
|
| 1895 |
+
"grad_norm": 0.6950060725212097,
|
| 1896 |
+
"learning_rate": 7.589352687812797e-05,
|
| 1897 |
+
"loss": 1.3658,
|
| 1898 |
+
"step": 1315
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 2.11,
|
| 1902 |
+
"grad_norm": 1.0462900400161743,
|
| 1903 |
+
"learning_rate": 7.467595522850805e-05,
|
| 1904 |
+
"loss": 1.4803,
|
| 1905 |
+
"step": 1320
|
| 1906 |
+
},
|
| 1907 |
+
{
|
| 1908 |
+
"epoch": 2.11,
|
| 1909 |
+
"grad_norm": 0.9355428218841553,
|
| 1910 |
+
"learning_rate": 7.346498422277214e-05,
|
| 1911 |
+
"loss": 1.5474,
|
| 1912 |
+
"step": 1325
|
| 1913 |
+
},
|
| 1914 |
+
{
|
| 1915 |
+
"epoch": 2.12,
|
| 1916 |
+
"grad_norm": 0.8697147369384766,
|
| 1917 |
+
"learning_rate": 7.22607199782747e-05,
|
| 1918 |
+
"loss": 1.4883,
|
| 1919 |
+
"step": 1330
|
| 1920 |
+
},
|
| 1921 |
+
{
|
| 1922 |
+
"epoch": 2.13,
|
| 1923 |
+
"grad_norm": 0.9624801874160767,
|
| 1924 |
+
"learning_rate": 7.106326802465692e-05,
|
| 1925 |
+
"loss": 1.3623,
|
| 1926 |
+
"step": 1335
|
| 1927 |
+
},
|
| 1928 |
+
{
|
| 1929 |
+
"epoch": 2.14,
|
| 1930 |
+
"grad_norm": 0.884113073348999,
|
| 1931 |
+
"learning_rate": 6.987273329459915e-05,
|
| 1932 |
+
"loss": 1.4287,
|
| 1933 |
+
"step": 1340
|
| 1934 |
+
},
|
| 1935 |
+
{
|
| 1936 |
+
"epoch": 2.15,
|
| 1937 |
+
"grad_norm": 0.8870747685432434,
|
| 1938 |
+
"learning_rate": 6.868922011462574e-05,
|
| 1939 |
+
"loss": 1.4426,
|
| 1940 |
+
"step": 1345
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 2.15,
|
| 1944 |
+
"grad_norm": 0.9415817260742188,
|
| 1945 |
+
"learning_rate": 6.751283219596263e-05,
|
| 1946 |
+
"loss": 1.4421,
|
| 1947 |
+
"step": 1350
|
| 1948 |
+
},
|
| 1949 |
+
{
|
| 1950 |
+
"epoch": 2.16,
|
| 1951 |
+
"grad_norm": 0.8890645503997803,
|
| 1952 |
+
"learning_rate": 6.634367262544956e-05,
|
| 1953 |
+
"loss": 1.4436,
|
| 1954 |
+
"step": 1355
|
| 1955 |
+
},
|
| 1956 |
+
{
|
| 1957 |
+
"epoch": 2.17,
|
| 1958 |
+
"grad_norm": 0.8797799944877625,
|
| 1959 |
+
"learning_rate": 6.518184385650595e-05,
|
| 1960 |
+
"loss": 1.3414,
|
| 1961 |
+
"step": 1360
|
| 1962 |
+
},
|
| 1963 |
+
{
|
| 1964 |
+
"epoch": 2.18,
|
| 1965 |
+
"grad_norm": 0.8096060156822205,
|
| 1966 |
+
"learning_rate": 6.402744770015371e-05,
|
| 1967 |
+
"loss": 1.3585,
|
| 1968 |
+
"step": 1365
|
| 1969 |
+
},
|
| 1970 |
+
{
|
| 1971 |
+
"epoch": 2.19,
|
| 1972 |
+
"grad_norm": 0.9265967607498169,
|
| 1973 |
+
"learning_rate": 6.28805853160948e-05,
|
| 1974 |
+
"loss": 1.4575,
|
| 1975 |
+
"step": 1370
|
| 1976 |
+
},
|
| 1977 |
+
{
|
| 1978 |
+
"epoch": 2.19,
|
| 1979 |
+
"grad_norm": 0.7921759486198425,
|
| 1980 |
+
"learning_rate": 6.174135720384715e-05,
|
| 1981 |
+
"loss": 1.5079,
|
| 1982 |
+
"step": 1375
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 2.2,
|
| 1986 |
+
"grad_norm": 0.9000433087348938,
|
| 1987 |
+
"learning_rate": 6.0609863193937554e-05,
|
| 1988 |
+
"loss": 1.4363,
|
| 1989 |
+
"step": 1380
|
| 1990 |
+
},
|
| 1991 |
+
{
|
| 1992 |
+
"epoch": 2.21,
|
| 1993 |
+
"grad_norm": 0.9442440271377563,
|
| 1994 |
+
"learning_rate": 5.9486202439153636e-05,
|
| 1995 |
+
"loss": 1.3952,
|
| 1996 |
+
"step": 1385
|
| 1997 |
+
},
|
| 1998 |
+
{
|
| 1999 |
+
"epoch": 2.22,
|
| 2000 |
+
"grad_norm": 0.9569089412689209,
|
| 2001 |
+
"learning_rate": 5.837047340585508e-05,
|
| 2002 |
+
"loss": 1.3183,
|
| 2003 |
+
"step": 1390
|
| 2004 |
+
},
|
| 2005 |
+
{
|
| 2006 |
+
"epoch": 2.23,
|
| 2007 |
+
"grad_norm": 0.7584913969039917,
|
| 2008 |
+
"learning_rate": 5.72627738653452e-05,
|
| 2009 |
+
"loss": 1.3542,
|
| 2010 |
+
"step": 1395
|
| 2011 |
+
},
|
| 2012 |
+
{
|
| 2013 |
+
"epoch": 2.23,
|
| 2014 |
+
"grad_norm": 0.9052013754844666,
|
| 2015 |
+
"learning_rate": 5.616320088530287e-05,
|
| 2016 |
+
"loss": 1.4462,
|
| 2017 |
+
"step": 1400
|
| 2018 |
+
},
|
| 2019 |
+
{
|
| 2020 |
+
"epoch": 2.23,
|
| 2021 |
+
"eval_loss": 1.6514110565185547,
|
| 2022 |
+
"eval_runtime": 53.4421,
|
| 2023 |
+
"eval_samples_per_second": 9.057,
|
| 2024 |
+
"eval_steps_per_second": 2.264,
|
| 2025 |
+
"step": 1400
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 2.24,
|
| 2029 |
+
"grad_norm": 0.9608349800109863,
|
| 2030 |
+
"learning_rate": 5.50718508212769e-05,
|
| 2031 |
+
"loss": 1.4392,
|
| 2032 |
+
"step": 1405
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 2.25,
|
| 2036 |
+
"grad_norm": 1.0080007314682007,
|
| 2037 |
+
"learning_rate": 5.398881930824208e-05,
|
| 2038 |
+
"loss": 1.4975,
|
| 2039 |
+
"step": 1410
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 2.26,
|
| 2043 |
+
"grad_norm": 0.9581565260887146,
|
| 2044 |
+
"learning_rate": 5.2914201252218776e-05,
|
| 2045 |
+
"loss": 1.3116,
|
| 2046 |
+
"step": 1415
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 2.27,
|
| 2050 |
+
"grad_norm": 0.9258535504341125,
|
| 2051 |
+
"learning_rate": 5.184809082195656e-05,
|
| 2052 |
+
"loss": 1.5025,
|
| 2053 |
+
"step": 1420
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 2.27,
|
| 2057 |
+
"grad_norm": 0.815497100353241,
|
| 2058 |
+
"learning_rate": 5.079058144068175e-05,
|
| 2059 |
+
"loss": 1.4159,
|
| 2060 |
+
"step": 1425
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 2.28,
|
| 2064 |
+
"grad_norm": 0.867760181427002,
|
| 2065 |
+
"learning_rate": 4.974176577791123e-05,
|
| 2066 |
+
"loss": 1.4266,
|
| 2067 |
+
"step": 1430
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 2.29,
|
| 2071 |
+
"grad_norm": 0.8949117660522461,
|
| 2072 |
+
"learning_rate": 4.870173574133139e-05,
|
| 2073 |
+
"loss": 1.4693,
|
| 2074 |
+
"step": 1435
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 2.3,
|
| 2078 |
+
"grad_norm": 0.9198044538497925,
|
| 2079 |
+
"learning_rate": 4.767058246874453e-05,
|
| 2080 |
+
"loss": 1.4362,
|
| 2081 |
+
"step": 1440
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 2.31,
|
| 2085 |
+
"grad_norm": 0.928459882736206,
|
| 2086 |
+
"learning_rate": 4.664839632008233e-05,
|
| 2087 |
+
"loss": 1.4778,
|
| 2088 |
+
"step": 1445
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 2.31,
|
| 2092 |
+
"grad_norm": 0.789795458316803,
|
| 2093 |
+
"learning_rate": 4.563526686948776e-05,
|
| 2094 |
+
"loss": 1.3265,
|
| 2095 |
+
"step": 1450
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 2.32,
|
| 2099 |
+
"grad_norm": 0.9109902381896973,
|
| 2100 |
+
"learning_rate": 4.4631282897465504e-05,
|
| 2101 |
+
"loss": 1.4303,
|
| 2102 |
+
"step": 1455
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 2.33,
|
| 2106 |
+
"grad_norm": 0.9469901919364929,
|
| 2107 |
+
"learning_rate": 4.363653238310223e-05,
|
| 2108 |
+
"loss": 1.3995,
|
| 2109 |
+
"step": 1460
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 2.34,
|
| 2113 |
+
"grad_norm": 0.8550880551338196,
|
| 2114 |
+
"learning_rate": 4.2651102496356985e-05,
|
| 2115 |
+
"loss": 1.3542,
|
| 2116 |
+
"step": 1465
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 2.35,
|
| 2120 |
+
"grad_norm": 0.8234253525733948,
|
| 2121 |
+
"learning_rate": 4.1675079590422553e-05,
|
| 2122 |
+
"loss": 1.3771,
|
| 2123 |
+
"step": 1470
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 2.35,
|
| 2127 |
+
"grad_norm": 0.9473711848258972,
|
| 2128 |
+
"learning_rate": 4.070854919415811e-05,
|
| 2129 |
+
"loss": 1.3968,
|
| 2130 |
+
"step": 1475
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 2.36,
|
| 2134 |
+
"grad_norm": 1.0804728269577026,
|
| 2135 |
+
"learning_rate": 3.9751596004594696e-05,
|
| 2136 |
+
"loss": 1.4282,
|
| 2137 |
+
"step": 1480
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 2.37,
|
| 2141 |
+
"grad_norm": 0.9122664332389832,
|
| 2142 |
+
"learning_rate": 3.88043038795128e-05,
|
| 2143 |
+
"loss": 1.4415,
|
| 2144 |
+
"step": 1485
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 2.38,
|
| 2148 |
+
"grad_norm": 0.851219892501831,
|
| 2149 |
+
"learning_rate": 3.786675583009419e-05,
|
| 2150 |
+
"loss": 1.3758,
|
| 2151 |
+
"step": 1490
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 2.39,
|
| 2155 |
+
"grad_norm": 0.8568178415298462,
|
| 2156 |
+
"learning_rate": 3.6939034013647516e-05,
|
| 2157 |
+
"loss": 1.3924,
|
| 2158 |
+
"step": 1495
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 2.39,
|
| 2162 |
+
"grad_norm": 0.8509604334831238,
|
| 2163 |
+
"learning_rate": 3.6021219726409064e-05,
|
| 2164 |
+
"loss": 1.3347,
|
| 2165 |
+
"step": 1500
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 2.4,
|
| 2169 |
+
"grad_norm": 0.9844736456871033,
|
| 2170 |
+
"learning_rate": 3.51133933964184e-05,
|
| 2171 |
+
"loss": 1.3425,
|
| 2172 |
+
"step": 1505
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 2.41,
|
| 2176 |
+
"grad_norm": 0.8552778959274292,
|
| 2177 |
+
"learning_rate": 3.421563457647091e-05,
|
| 2178 |
+
"loss": 1.4085,
|
| 2179 |
+
"step": 1510
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 2.42,
|
| 2183 |
+
"grad_norm": 0.8389209508895874,
|
| 2184 |
+
"learning_rate": 3.3328021937146175e-05,
|
| 2185 |
+
"loss": 1.4795,
|
| 2186 |
+
"step": 1515
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 2.43,
|
| 2190 |
+
"grad_norm": 0.7627727389335632,
|
| 2191 |
+
"learning_rate": 3.2450633259914324e-05,
|
| 2192 |
+
"loss": 1.3852,
|
| 2193 |
+
"step": 1520
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 2.43,
|
| 2197 |
+
"grad_norm": 0.7493878602981567,
|
| 2198 |
+
"learning_rate": 3.158354543032004e-05,
|
| 2199 |
+
"loss": 1.4683,
|
| 2200 |
+
"step": 1525
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 2.44,
|
| 2204 |
+
"grad_norm": 0.8921322226524353,
|
| 2205 |
+
"learning_rate": 3.0726834431244916e-05,
|
| 2206 |
+
"loss": 1.3623,
|
| 2207 |
+
"step": 1530
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 2.45,
|
| 2211 |
+
"grad_norm": 0.934528648853302,
|
| 2212 |
+
"learning_rate": 2.9880575336249347e-05,
|
| 2213 |
+
"loss": 1.4292,
|
| 2214 |
+
"step": 1535
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 2.46,
|
| 2218 |
+
"grad_norm": 0.9410387873649597,
|
| 2219 |
+
"learning_rate": 2.904484230299341e-05,
|
| 2220 |
+
"loss": 1.3741,
|
| 2221 |
+
"step": 1540
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 2.47,
|
| 2225 |
+
"grad_norm": 0.8113580942153931,
|
| 2226 |
+
"learning_rate": 2.821970856673898e-05,
|
| 2227 |
+
"loss": 1.3496,
|
| 2228 |
+
"step": 1545
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 2.47,
|
| 2232 |
+
"grad_norm": 0.822819709777832,
|
| 2233 |
+
"learning_rate": 2.7405246433931603e-05,
|
| 2234 |
+
"loss": 1.387,
|
| 2235 |
+
"step": 1550
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 2.48,
|
| 2239 |
+
"grad_norm": 0.7705480456352234,
|
| 2240 |
+
"learning_rate": 2.6601527275864743e-05,
|
| 2241 |
+
"loss": 1.4588,
|
| 2242 |
+
"step": 1555
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 2.49,
|
| 2246 |
+
"grad_norm": 0.8231070637702942,
|
| 2247 |
+
"learning_rate": 2.5808621522425134e-05,
|
| 2248 |
+
"loss": 1.5636,
|
| 2249 |
+
"step": 1560
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 2.5,
|
| 2253 |
+
"grad_norm": 0.8312299251556396,
|
| 2254 |
+
"learning_rate": 2.5026598655921182e-05,
|
| 2255 |
+
"loss": 1.4001,
|
| 2256 |
+
"step": 1565
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 2.5,
|
| 2260 |
+
"grad_norm": 0.7779568433761597,
|
| 2261 |
+
"learning_rate": 2.4255527204994268e-05,
|
| 2262 |
+
"loss": 1.4055,
|
| 2263 |
+
"step": 1570
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 2.51,
|
| 2267 |
+
"grad_norm": 0.9976507425308228,
|
| 2268 |
+
"learning_rate": 2.3495474738613595e-05,
|
| 2269 |
+
"loss": 1.4549,
|
| 2270 |
+
"step": 1575
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 2.52,
|
| 2274 |
+
"grad_norm": 0.9732272624969482,
|
| 2275 |
+
"learning_rate": 2.274650786015491e-05,
|
| 2276 |
+
"loss": 1.3406,
|
| 2277 |
+
"step": 1580
|
| 2278 |
+
},
|
| 2279 |
+
{
|
| 2280 |
+
"epoch": 2.53,
|
| 2281 |
+
"grad_norm": 0.8873571753501892,
|
| 2282 |
+
"learning_rate": 2.200869220156439e-05,
|
| 2283 |
+
"loss": 1.3103,
|
| 2284 |
+
"step": 1585
|
| 2285 |
+
},
|
| 2286 |
+
{
|
| 2287 |
+
"epoch": 2.54,
|
| 2288 |
+
"grad_norm": 0.940963625907898,
|
| 2289 |
+
"learning_rate": 2.1282092417606998e-05,
|
| 2290 |
+
"loss": 1.3055,
|
| 2291 |
+
"step": 1590
|
| 2292 |
+
},
|
| 2293 |
+
{
|
| 2294 |
+
"epoch": 2.54,
|
| 2295 |
+
"grad_norm": 0.9407901167869568,
|
| 2296 |
+
"learning_rate": 2.0566772180201002e-05,
|
| 2297 |
+
"loss": 1.3727,
|
| 2298 |
+
"step": 1595
|
| 2299 |
+
},
|
| 2300 |
+
{
|
| 2301 |
+
"epoch": 2.55,
|
| 2302 |
+
"grad_norm": 0.7993718981742859,
|
| 2303 |
+
"learning_rate": 1.986279417283823e-05,
|
| 2304 |
+
"loss": 1.3236,
|
| 2305 |
+
"step": 1600
|
| 2306 |
+
},
|
| 2307 |
+
{
|
| 2308 |
+
"epoch": 2.55,
|
| 2309 |
+
"eval_loss": 1.6478137969970703,
|
| 2310 |
+
"eval_runtime": 53.9373,
|
| 2311 |
+
"eval_samples_per_second": 8.973,
|
| 2312 |
+
"eval_steps_per_second": 2.243,
|
| 2313 |
+
"step": 1600
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 2.56,
|
| 2317 |
+
"grad_norm": 0.9243110418319702,
|
| 2318 |
+
"learning_rate": 1.9170220085091354e-05,
|
| 2319 |
+
"loss": 1.3978,
|
| 2320 |
+
"step": 1605
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 2.57,
|
| 2324 |
+
"grad_norm": 0.9336215257644653,
|
| 2325 |
+
"learning_rate": 1.8489110607207796e-05,
|
| 2326 |
+
"loss": 1.3513,
|
| 2327 |
+
"step": 1610
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 2.58,
|
| 2331 |
+
"grad_norm": 0.9775585532188416,
|
| 2332 |
+
"learning_rate": 1.7819525424791637e-05,
|
| 2333 |
+
"loss": 1.3751,
|
| 2334 |
+
"step": 1615
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 2.58,
|
| 2338 |
+
"grad_norm": 0.8599871397018433,
|
| 2339 |
+
"learning_rate": 1.7161523213573115e-05,
|
| 2340 |
+
"loss": 1.4563,
|
| 2341 |
+
"step": 1620
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 2.59,
|
| 2345 |
+
"grad_norm": 0.8650587797164917,
|
| 2346 |
+
"learning_rate": 1.6515161634267216e-05,
|
| 2347 |
+
"loss": 1.3313,
|
| 2348 |
+
"step": 1625
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 2.6,
|
| 2352 |
+
"grad_norm": 0.8923302888870239,
|
| 2353 |
+
"learning_rate": 1.5880497327520546e-05,
|
| 2354 |
+
"loss": 1.4745,
|
| 2355 |
+
"step": 1630
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 2.61,
|
| 2359 |
+
"grad_norm": 0.9115777611732483,
|
| 2360 |
+
"learning_rate": 1.5257585908948172e-05,
|
| 2361 |
+
"loss": 1.5308,
|
| 2362 |
+
"step": 1635
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 2.62,
|
| 2366 |
+
"grad_norm": 0.8948424458503723,
|
| 2367 |
+
"learning_rate": 1.464648196425981e-05,
|
| 2368 |
+
"loss": 1.4087,
|
| 2369 |
+
"step": 1640
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 2.62,
|
| 2373 |
+
"grad_norm": 0.9226679801940918,
|
| 2374 |
+
"learning_rate": 1.4047239044476594e-05,
|
| 2375 |
+
"loss": 1.3918,
|
| 2376 |
+
"step": 1645
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 2.63,
|
| 2380 |
+
"grad_norm": 0.8747596144676208,
|
| 2381 |
+
"learning_rate": 1.345990966123846e-05,
|
| 2382 |
+
"loss": 1.3892,
|
| 2383 |
+
"step": 1650
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 2.64,
|
| 2387 |
+
"grad_norm": 0.8658227324485779,
|
| 2388 |
+
"learning_rate": 1.288454528220238e-05,
|
| 2389 |
+
"loss": 1.4378,
|
| 2390 |
+
"step": 1655
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 2.65,
|
| 2394 |
+
"grad_norm": 0.8440380096435547,
|
| 2395 |
+
"learning_rate": 1.2321196326532428e-05,
|
| 2396 |
+
"loss": 1.3919,
|
| 2397 |
+
"step": 1660
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 2.66,
|
| 2401 |
+
"grad_norm": 0.8718552589416504,
|
| 2402 |
+
"learning_rate": 1.176991216048141e-05,
|
| 2403 |
+
"loss": 1.3446,
|
| 2404 |
+
"step": 1665
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 2.66,
|
| 2408 |
+
"grad_norm": 1.080561876296997,
|
| 2409 |
+
"learning_rate": 1.123074109306501e-05,
|
| 2410 |
+
"loss": 1.5042,
|
| 2411 |
+
"step": 1670
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 2.67,
|
| 2415 |
+
"grad_norm": 0.9557725191116333,
|
| 2416 |
+
"learning_rate": 1.070373037182839e-05,
|
| 2417 |
+
"loss": 1.4899,
|
| 2418 |
+
"step": 1675
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 2.68,
|
| 2422 |
+
"grad_norm": 0.8951625823974609,
|
| 2423 |
+
"learning_rate": 1.0188926178705976e-05,
|
| 2424 |
+
"loss": 1.3954,
|
| 2425 |
+
"step": 1680
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 2.69,
|
| 2429 |
+
"grad_norm": 0.8350385427474976,
|
| 2430 |
+
"learning_rate": 9.686373625974398e-06,
|
| 2431 |
+
"loss": 1.3416,
|
| 2432 |
+
"step": 1685
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 2.7,
|
| 2436 |
+
"grad_norm": 0.7895828485488892,
|
| 2437 |
+
"learning_rate": 9.196116752299504e-06,
|
| 2438 |
+
"loss": 1.4138,
|
| 2439 |
+
"step": 1690
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 2.7,
|
| 2443 |
+
"grad_norm": 0.8298912644386292,
|
| 2444 |
+
"learning_rate": 8.718198518877063e-06,
|
| 2445 |
+
"loss": 1.4092,
|
| 2446 |
+
"step": 1695
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 2.71,
|
| 2450 |
+
"grad_norm": 0.7747901082038879,
|
| 2451 |
+
"learning_rate": 8.252660805668172e-06,
|
| 2452 |
+
"loss": 1.4483,
|
| 2453 |
+
"step": 1700
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 2.72,
|
| 2457 |
+
"grad_norm": 0.9418133497238159,
|
| 2458 |
+
"learning_rate": 7.799544407729247e-06,
|
| 2459 |
+
"loss": 1.2875,
|
| 2460 |
+
"step": 1705
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 2.73,
|
| 2464 |
+
"grad_norm": 0.8848152756690979,
|
| 2465 |
+
"learning_rate": 7.3588890316373e-06,
|
| 2466 |
+
"loss": 1.3768,
|
| 2467 |
+
"step": 1710
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 2.74,
|
| 2471 |
+
"grad_norm": 0.8357783555984497,
|
| 2472 |
+
"learning_rate": 6.930733292010299e-06,
|
| 2473 |
+
"loss": 1.2686,
|
| 2474 |
+
"step": 1715
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 2.74,
|
| 2478 |
+
"grad_norm": 0.8660513758659363,
|
| 2479 |
+
"learning_rate": 6.5151147081234314e-06,
|
| 2480 |
+
"loss": 1.5152,
|
| 2481 |
+
"step": 1720
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 2.75,
|
| 2485 |
+
"grad_norm": 0.9291744232177734,
|
| 2486 |
+
"learning_rate": 6.112069700621247e-06,
|
| 2487 |
+
"loss": 1.3963,
|
| 2488 |
+
"step": 1725
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 2.76,
|
| 2492 |
+
"grad_norm": 0.9582973718643188,
|
| 2493 |
+
"learning_rate": 5.721633588326263e-06,
|
| 2494 |
+
"loss": 1.3672,
|
| 2495 |
+
"step": 1730
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 2.77,
|
| 2499 |
+
"grad_norm": 0.8532952070236206,
|
| 2500 |
+
"learning_rate": 5.343840585143755e-06,
|
| 2501 |
+
"loss": 1.4313,
|
| 2502 |
+
"step": 1735
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 2.78,
|
| 2506 |
+
"grad_norm": 0.8306805491447449,
|
| 2507 |
+
"learning_rate": 4.9787237970638046e-06,
|
| 2508 |
+
"loss": 1.348,
|
| 2509 |
+
"step": 1740
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 2.78,
|
| 2513 |
+
"grad_norm": 0.9041390419006348,
|
| 2514 |
+
"learning_rate": 4.626315219260074e-06,
|
| 2515 |
+
"loss": 1.323,
|
| 2516 |
+
"step": 1745
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 2.79,
|
| 2520 |
+
"grad_norm": 0.9139644503593445,
|
| 2521 |
+
"learning_rate": 4.2866457332861095e-06,
|
| 2522 |
+
"loss": 1.4489,
|
| 2523 |
+
"step": 1750
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 2.8,
|
| 2527 |
+
"grad_norm": 0.7889164090156555,
|
| 2528 |
+
"learning_rate": 3.959745104369227e-06,
|
| 2529 |
+
"loss": 1.2446,
|
| 2530 |
+
"step": 1755
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 2.81,
|
| 2534 |
+
"grad_norm": 0.8802595734596252,
|
| 2535 |
+
"learning_rate": 3.645641978802183e-06,
|
| 2536 |
+
"loss": 1.3255,
|
| 2537 |
+
"step": 1760
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 2.82,
|
| 2541 |
+
"grad_norm": 0.8217524290084839,
|
| 2542 |
+
"learning_rate": 3.3443638814328765e-06,
|
| 2543 |
+
"loss": 1.3215,
|
| 2544 |
+
"step": 1765
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 2.82,
|
| 2548 |
+
"grad_norm": 1.0614471435546875,
|
| 2549 |
+
"learning_rate": 3.0559372132523354e-06,
|
| 2550 |
+
"loss": 1.5207,
|
| 2551 |
+
"step": 1770
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 2.83,
|
| 2555 |
+
"grad_norm": 0.8429582715034485,
|
| 2556 |
+
"learning_rate": 2.7803872490812287e-06,
|
| 2557 |
+
"loss": 1.3383,
|
| 2558 |
+
"step": 1775
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 2.84,
|
| 2562 |
+
"grad_norm": 0.8786703944206238,
|
| 2563 |
+
"learning_rate": 2.5177381353549907e-06,
|
| 2564 |
+
"loss": 1.3797,
|
| 2565 |
+
"step": 1780
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 2.85,
|
| 2569 |
+
"grad_norm": 0.9570217132568359,
|
| 2570 |
+
"learning_rate": 2.268012888007947e-06,
|
| 2571 |
+
"loss": 1.3989,
|
| 2572 |
+
"step": 1785
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 2.86,
|
| 2576 |
+
"grad_norm": 0.8841099739074707,
|
| 2577 |
+
"learning_rate": 2.0312333904563605e-06,
|
| 2578 |
+
"loss": 1.3689,
|
| 2579 |
+
"step": 1790
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 2.86,
|
| 2583 |
+
"grad_norm": 0.9913931488990784,
|
| 2584 |
+
"learning_rate": 1.8074203916808472e-06,
|
| 2585 |
+
"loss": 1.4612,
|
| 2586 |
+
"step": 1795
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 2.87,
|
| 2590 |
+
"grad_norm": 0.84140545129776,
|
| 2591 |
+
"learning_rate": 1.5965935044080003e-06,
|
| 2592 |
+
"loss": 1.4728,
|
| 2593 |
+
"step": 1800
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 2.87,
|
| 2597 |
+
"eval_loss": 1.645330786705017,
|
| 2598 |
+
"eval_runtime": 53.8569,
|
| 2599 |
+
"eval_samples_per_second": 8.987,
|
| 2600 |
+
"eval_steps_per_second": 2.247,
|
| 2601 |
+
"step": 1800
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 2.88,
|
| 2605 |
+
"grad_norm": 1.0088138580322266,
|
| 2606 |
+
"learning_rate": 1.3987712033919796e-06,
|
| 2607 |
+
"loss": 1.4539,
|
| 2608 |
+
"step": 1805
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 2.89,
|
| 2612 |
+
"grad_norm": 0.853320837020874,
|
| 2613 |
+
"learning_rate": 1.2139708237953073e-06,
|
| 2614 |
+
"loss": 1.3314,
|
| 2615 |
+
"step": 1810
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 2.9,
|
| 2619 |
+
"grad_norm": 0.9394313097000122,
|
| 2620 |
+
"learning_rate": 1.0422085596699658e-06,
|
| 2621 |
+
"loss": 1.4403,
|
| 2622 |
+
"step": 1815
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 2.9,
|
| 2626 |
+
"grad_norm": 0.8725711107254028,
|
| 2627 |
+
"learning_rate": 8.834994625382341e-07,
|
| 2628 |
+
"loss": 1.4662,
|
| 2629 |
+
"step": 1820
|
| 2630 |
+
},
|
| 2631 |
+
{
|
| 2632 |
+
"epoch": 2.91,
|
| 2633 |
+
"grad_norm": 0.9195972084999084,
|
| 2634 |
+
"learning_rate": 7.378574400737414e-07,
|
| 2635 |
+
"loss": 1.3785,
|
| 2636 |
+
"step": 1825
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 2.92,
|
| 2640 |
+
"grad_norm": 0.9402215480804443,
|
| 2641 |
+
"learning_rate": 6.052952548827428e-07,
|
| 2642 |
+
"loss": 1.4565,
|
| 2643 |
+
"step": 1830
|
| 2644 |
+
},
|
| 2645 |
+
{
|
| 2646 |
+
"epoch": 2.93,
|
| 2647 |
+
"grad_norm": 0.8402209281921387,
|
| 2648 |
+
"learning_rate": 4.858245233857305e-07,
|
| 2649 |
+
"loss": 1.371,
|
| 2650 |
+
"step": 1835
|
| 2651 |
+
},
|
| 2652 |
+
{
|
| 2653 |
+
"epoch": 2.94,
|
| 2654 |
+
"grad_norm": 0.7974345088005066,
|
| 2655 |
+
"learning_rate": 3.7945571479948147e-07,
|
| 2656 |
+
"loss": 1.3159,
|
| 2657 |
+
"step": 1840
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"epoch": 2.94,
|
| 2661 |
+
"grad_norm": 0.8706817626953125,
|
| 2662 |
+
"learning_rate": 2.8619815021964085e-07,
|
| 2663 |
+
"loss": 1.521,
|
| 2664 |
+
"step": 1845
|
| 2665 |
+
},
|
| 2666 |
+
{
|
| 2667 |
+
"epoch": 2.95,
|
| 2668 |
+
"grad_norm": 0.9706933498382568,
|
| 2669 |
+
"learning_rate": 2.0606000180390935e-07,
|
| 2670 |
+
"loss": 1.3561,
|
| 2671 |
+
"step": 1850
|
| 2672 |
+
},
|
| 2673 |
+
{
|
| 2674 |
+
"epoch": 2.96,
|
| 2675 |
+
"grad_norm": 0.9271072149276733,
|
| 2676 |
+
"learning_rate": 1.3904829205594882e-07,
|
| 2677 |
+
"loss": 1.4329,
|
| 2678 |
+
"step": 1855
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 2.97,
|
| 2682 |
+
"grad_norm": 0.8780162930488586,
|
| 2683 |
+
"learning_rate": 8.516889320995813e-08,
|
| 2684 |
+
"loss": 1.3876,
|
| 2685 |
+
"step": 1860
|
| 2686 |
+
},
|
| 2687 |
+
{
|
| 2688 |
+
"epoch": 2.98,
|
| 2689 |
+
"grad_norm": 0.8870643377304077,
|
| 2690 |
+
"learning_rate": 4.442652671611813e-08,
|
| 2691 |
+
"loss": 1.4097,
|
| 2692 |
+
"step": 1865
|
| 2693 |
+
},
|
| 2694 |
+
{
|
| 2695 |
+
"epoch": 2.98,
|
| 2696 |
+
"grad_norm": 0.9191935062408447,
|
| 2697 |
+
"learning_rate": 1.6824762826822546e-08,
|
| 2698 |
+
"loss": 1.5358,
|
| 2699 |
+
"step": 1870
|
| 2700 |
+
},
|
| 2701 |
+
{
|
| 2702 |
+
"epoch": 2.99,
|
| 2703 |
+
"grad_norm": 0.856283962726593,
|
| 2704 |
+
"learning_rate": 2.3660202838615695e-09,
|
| 2705 |
+
"loss": 1.407,
|
| 2706 |
+
"step": 1875
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 3.0,
|
| 2710 |
+
"step": 1878,
|
| 2711 |
+
"total_flos": 2.5822124463489024e+17,
|
| 2712 |
+
"train_loss": 1.5555958059776072,
|
| 2713 |
+
"train_runtime": 10085.6682,
|
| 2714 |
+
"train_samples_per_second": 2.982,
|
| 2715 |
+
"train_steps_per_second": 0.186
|
| 2716 |
+
}
|
| 2717 |
+
],
|
| 2718 |
+
"logging_steps": 5,
|
| 2719 |
+
"max_steps": 1878,
|
| 2720 |
+
"num_input_tokens_seen": 0,
|
| 2721 |
+
"num_train_epochs": 3,
|
| 2722 |
+
"save_steps": 400,
|
| 2723 |
+
"total_flos": 2.5822124463489024e+17,
|
| 2724 |
+
"train_batch_size": 4,
|
| 2725 |
+
"trial_name": null,
|
| 2726 |
+
"trial_params": null
|
| 2727 |
+
}
|
lora_8bit_adapter_raw/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51aec8d076aef3caca86c6c9565fcad6f6ef188a64bee4f6062684d898d4ed76
|
| 3 |
+
size 5048
|
lora_8bit_adapter_raw/training_eval_loss.png
ADDED
|
lora_8bit_adapter_raw/training_loss.png
ADDED
|