Upload folder using huggingface_hub
Browse files- added_tokens.json +6 -0
- config.json +1 -0
- generation_config.json +14 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- params.txt +45 -0
- preprocessor_config.json +24 -0
- special_tokens_map.json +20 -0
- tokenizer.json +0 -0
- tokenizer_config.json +53 -0
- trainer_state.json +933 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<image>": 151646,
|
| 3 |
+
"<|endoftext|>": 151643,
|
| 4 |
+
"<|im_end|>": 151645,
|
| 5 |
+
"<|im_start|>": 151644
|
| 6 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"_name_or_path": "lmms-lab/llava-onevision-qwen2-0.5b-ov", "architectures": ["LlavaQwenForCausalLM"], "attention_dropout": 0.0, "bos_token_id": 151643, "eos_token_id": 151645, "hidden_act": "silu", "hidden_size": 896, "ignore_index": -100, "image_aspect_ratio": "anyres_max_9", "image_crop_resolution": null, "image_grid_pinpoints": [[384, 384], [384, 768], [384, 1152], [384, 1536], [384, 1920], [384, 2304], [768, 384], [768, 768], [768, 1152], [768, 1536], [768, 1920], [768, 2304], [1152, 384], [1152, 768], [1152, 1152], [1152, 1536], [1152, 1920], [1152, 2304], [1536, 384], [1536, 768], [1536, 1152], [1536, 1536], [1536, 1920], [1536, 2304], [1920, 384], [1920, 768], [1920, 1152], [1920, 1536], [1920, 1920], [1920, 2304], [2304, 384], [2304, 768], [2304, 1152], [2304, 1536], [2304, 1920], [2304, 2304]], "image_split_resolution": null, "image_token_index": 151646, "initializer_range": 0.02, "intermediate_size": 4864, "max_position_embeddings": 32768, "max_window_layers": 24, "mm_hidden_size": 1152, "mm_newline_position": "one_token", "mm_patch_merge_type": "spatial_unpad", "mm_projector_lr": null, "mm_projector_type": "mlp2x_gelu", "mm_resampler_type": null, "mm_spatial_pool_mode": "bilinear", "mm_tunable_parts": "mm_language_model", "mm_use_im_patch_token": false, "mm_use_im_start_end": false, "mm_vision_select_feature": "patch", "mm_vision_select_layer": -2, "mm_vision_tower": "google/siglip-so400m-patch14-384", "mm_vision_tower_lr": 2e-06, "model_type": "llava", "num_attention_heads": 14, "num_hidden_layers": 24, "num_key_value_heads": 2, "pos_skipping_range": 4096, "projector_hidden_act": "gelu", "rms_norm_eps": 1e-06, "rope_scaling": null, "rope_theta": 1000000.0, "sliding_window": 32768, "text_config": {"model_type": "llama"}, "tokenizer_model_max_length": 32768, "tokenizer_padding_side": "right", "torch_dtype": "bfloat16", "transformers_version": "4.40.0.dev0", "use_cache": true, "use_mm_proj": true, "use_pos_skipping": false, "use_sliding_window": false, "vision_config": {"hidden_size": 1024, "image_size": 336, "intermediate_size": 4096, "model_type": "clip_vision_model", "num_attention_heads": 16, "num_hidden_layers": 24, "patch_size": 14, "projection_dim": 768, "vocab_size": 32000}, "vision_feature_layer": -2, "vision_feature_select_strategy": "default", "vision_tower_pretrained": null, "vocab_size": 151936}
|
generation_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 151643,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
151645,
|
| 6 |
+
151643
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 151643,
|
| 9 |
+
"repetition_penalty": 1.1,
|
| 10 |
+
"temperature": 0.7,
|
| 11 |
+
"top_k": 20,
|
| 12 |
+
"top_p": 0.8,
|
| 13 |
+
"transformers_version": "4.40.0.dev0"
|
| 14 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea8b3527dcae5baae3e683081a25d08fa1ddf9fcda6e2b72d8355bb32030a952
|
| 3 |
+
size 1787335336
|
params.txt
ADDED
|
@@ -0,0 +1,45 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
train_old.py
|
| 2 |
+
deepspeed /workspace/llavaguard/configs/zero3.json
|
| 3 |
+
model_name_or_path lmms-lab/llava-onevision-qwen2-0.5b-ov
|
| 4 |
+
version qwen_1_5
|
| 5 |
+
data_path /common-repos/LlavaGuard/data/LlavaGuard-DS/v24/train_oversampled.json
|
| 6 |
+
data_path_eval /common-repos/LlavaGuard/data/LlavaGuard-DS/v24/eval.json
|
| 7 |
+
image_folder /common-repos
|
| 8 |
+
mm_tunable_parts=mm_language_model
|
| 9 |
+
mm_vision_tower_lr=2e-6
|
| 10 |
+
vision_tower google/siglip-so400m-patch14-384
|
| 11 |
+
mm_projector_type mlp2x_gelu
|
| 12 |
+
mm_vision_select_layer -2
|
| 13 |
+
mm_use_im_start_end False
|
| 14 |
+
mm_use_im_patch_token False
|
| 15 |
+
group_by_modality_length True
|
| 16 |
+
image_aspect_ratio anyres_max_9
|
| 17 |
+
image_grid_pinpoints (1x1),...,(6x6)
|
| 18 |
+
mm_patch_merge_type spatial_unpad
|
| 19 |
+
bf16 True
|
| 20 |
+
run_name LlavaGuard-v1.2-mini-ov-lmms-lab_llava-onevision-qwen2-0.5b-ov-LlavaGuard-DS-v24
|
| 21 |
+
output_dir /common-repos/LlavaGuard/models/LlavaGuard-v1.2-mini-ov/v24
|
| 22 |
+
num_train_epochs 3
|
| 23 |
+
per_device_train_batch_size 1
|
| 24 |
+
per_device_eval_batch_size 2
|
| 25 |
+
gradient_accumulation_steps 25
|
| 26 |
+
evaluation_strategy no
|
| 27 |
+
eval_steps 1
|
| 28 |
+
save_strategy epoch
|
| 29 |
+
save_steps 1
|
| 30 |
+
save_total_limit 1
|
| 31 |
+
learning_rate 1e-5
|
| 32 |
+
weight_decay 0.
|
| 33 |
+
warmup_ratio 0.03
|
| 34 |
+
lr_scheduler_type cosine
|
| 35 |
+
logging_steps 1
|
| 36 |
+
tf32 True
|
| 37 |
+
model_max_length 32768
|
| 38 |
+
gradient_checkpointing True
|
| 39 |
+
dataloader_num_workers 4
|
| 40 |
+
lazy_preprocess True
|
| 41 |
+
report_to wandb
|
| 42 |
+
torch_compile True
|
| 43 |
+
torch_compile_backend inductor
|
| 44 |
+
dataloader_drop_last True
|
| 45 |
+
Eval date: 22/11/2024 07:49:11
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"do_convert_rgb": null,
|
| 3 |
+
"do_normalize": true,
|
| 4 |
+
"do_rescale": true,
|
| 5 |
+
"do_resize": true,
|
| 6 |
+
"image_mean": [
|
| 7 |
+
0.5,
|
| 8 |
+
0.5,
|
| 9 |
+
0.5
|
| 10 |
+
],
|
| 11 |
+
"image_processor_type": "SiglipImageProcessor",
|
| 12 |
+
"image_std": [
|
| 13 |
+
0.5,
|
| 14 |
+
0.5,
|
| 15 |
+
0.5
|
| 16 |
+
],
|
| 17 |
+
"processor_class": "LlavaProcessor",
|
| 18 |
+
"resample": 3,
|
| 19 |
+
"rescale_factor": 0.00392156862745098,
|
| 20 |
+
"size": {
|
| 21 |
+
"height": 384,
|
| 22 |
+
"width": 384
|
| 23 |
+
}
|
| 24 |
+
}
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|im_start|>",
|
| 4 |
+
"<|im_end|>"
|
| 5 |
+
],
|
| 6 |
+
"eos_token": {
|
| 7 |
+
"content": "<|im_end|>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false
|
| 12 |
+
},
|
| 13 |
+
"pad_token": {
|
| 14 |
+
"content": "<|endoftext|>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false
|
| 19 |
+
}
|
| 20 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"151643": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"151644": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"151645": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"151646": {
|
| 29 |
+
"content": "<image>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"additional_special_tokens": [
|
| 38 |
+
"<|im_start|>",
|
| 39 |
+
"<|im_end|>"
|
| 40 |
+
],
|
| 41 |
+
"bos_token": null,
|
| 42 |
+
"chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 43 |
+
"clean_up_tokenization_spaces": false,
|
| 44 |
+
"eos_token": "<|im_end|>",
|
| 45 |
+
"errors": "replace",
|
| 46 |
+
"model_max_length": 32768,
|
| 47 |
+
"pad_token": "<|endoftext|>",
|
| 48 |
+
"padding_side": "right",
|
| 49 |
+
"processor_class": "LlavaProcessor",
|
| 50 |
+
"split_special_tokens": false,
|
| 51 |
+
"tokenizer_class": "Qwen2Tokenizer",
|
| 52 |
+
"unk_token": null
|
| 53 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,933 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9398359161349132,
|
| 5 |
+
"eval_steps": 1.0,
|
| 6 |
+
"global_step": 129,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 9.694531367351816,
|
| 14 |
+
"learning_rate": 0.0,
|
| 15 |
+
"loss": 0.9609,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.05,
|
| 20 |
+
"grad_norm": 6.0698578450559015,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.8693,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07,
|
| 27 |
+
"grad_norm": 6.5557897846625535,
|
| 28 |
+
"learning_rate": 7.924812503605782e-06,
|
| 29 |
+
"loss": 0.8571,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.09,
|
| 34 |
+
"grad_norm": 5.8249565635426235,
|
| 35 |
+
"learning_rate": 1e-05,
|
| 36 |
+
"loss": 0.8098,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.11,
|
| 41 |
+
"grad_norm": 7.157097194857861,
|
| 42 |
+
"learning_rate": 1e-05,
|
| 43 |
+
"loss": 0.7421,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.14,
|
| 48 |
+
"grad_norm": 6.7490063693246185,
|
| 49 |
+
"learning_rate": 1e-05,
|
| 50 |
+
"loss": 0.7184,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.16,
|
| 55 |
+
"grad_norm": 6.989582963023535,
|
| 56 |
+
"learning_rate": 1e-05,
|
| 57 |
+
"loss": 0.7101,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.18,
|
| 62 |
+
"grad_norm": 4.856325406983797,
|
| 63 |
+
"learning_rate": 1e-05,
|
| 64 |
+
"loss": 0.6815,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.21,
|
| 69 |
+
"grad_norm": 4.893753189326447,
|
| 70 |
+
"learning_rate": 1e-05,
|
| 71 |
+
"loss": 0.6975,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.23,
|
| 76 |
+
"grad_norm": 4.8564394287744745,
|
| 77 |
+
"learning_rate": 1e-05,
|
| 78 |
+
"loss": 0.65,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.25,
|
| 83 |
+
"grad_norm": 4.638777805285931,
|
| 84 |
+
"learning_rate": 1e-05,
|
| 85 |
+
"loss": 0.5817,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.27,
|
| 90 |
+
"grad_norm": 3.701804268792858,
|
| 91 |
+
"learning_rate": 1e-05,
|
| 92 |
+
"loss": 0.6005,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.3,
|
| 97 |
+
"grad_norm": 3.08244950250501,
|
| 98 |
+
"learning_rate": 1e-05,
|
| 99 |
+
"loss": 0.5529,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.32,
|
| 104 |
+
"grad_norm": 3.210541622720776,
|
| 105 |
+
"learning_rate": 1e-05,
|
| 106 |
+
"loss": 0.5654,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.34,
|
| 111 |
+
"grad_norm": 3.0384959005478867,
|
| 112 |
+
"learning_rate": 1e-05,
|
| 113 |
+
"loss": 0.5184,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.36,
|
| 118 |
+
"grad_norm": 2.6850393439015092,
|
| 119 |
+
"learning_rate": 1e-05,
|
| 120 |
+
"loss": 0.5234,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.39,
|
| 125 |
+
"grad_norm": 2.8596154120661748,
|
| 126 |
+
"learning_rate": 1e-05,
|
| 127 |
+
"loss": 0.5005,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.41,
|
| 132 |
+
"grad_norm": 2.726391191331952,
|
| 133 |
+
"learning_rate": 1e-05,
|
| 134 |
+
"loss": 0.4691,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.43,
|
| 139 |
+
"grad_norm": 2.934188314387199,
|
| 140 |
+
"learning_rate": 1e-05,
|
| 141 |
+
"loss": 0.5265,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.46,
|
| 146 |
+
"grad_norm": 2.496046815968877,
|
| 147 |
+
"learning_rate": 1e-05,
|
| 148 |
+
"loss": 0.46,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.48,
|
| 153 |
+
"grad_norm": 2.6527493698931703,
|
| 154 |
+
"learning_rate": 1e-05,
|
| 155 |
+
"loss": 0.4633,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.5,
|
| 160 |
+
"grad_norm": 2.3371216207823364,
|
| 161 |
+
"learning_rate": 1e-05,
|
| 162 |
+
"loss": 0.4512,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.52,
|
| 167 |
+
"grad_norm": 2.5029789662415576,
|
| 168 |
+
"learning_rate": 1e-05,
|
| 169 |
+
"loss": 0.4538,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.55,
|
| 174 |
+
"grad_norm": 2.3654775885807435,
|
| 175 |
+
"learning_rate": 1e-05,
|
| 176 |
+
"loss": 0.4503,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.57,
|
| 181 |
+
"grad_norm": 2.520897440850751,
|
| 182 |
+
"learning_rate": 1e-05,
|
| 183 |
+
"loss": 0.4393,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.59,
|
| 188 |
+
"grad_norm": 2.2973148314047998,
|
| 189 |
+
"learning_rate": 1e-05,
|
| 190 |
+
"loss": 0.4246,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.62,
|
| 195 |
+
"grad_norm": 2.064310434148654,
|
| 196 |
+
"learning_rate": 1e-05,
|
| 197 |
+
"loss": 0.4466,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.64,
|
| 202 |
+
"grad_norm": 2.1518579274793614,
|
| 203 |
+
"learning_rate": 1e-05,
|
| 204 |
+
"loss": 0.4157,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.66,
|
| 209 |
+
"grad_norm": 2.101663348817292,
|
| 210 |
+
"learning_rate": 1e-05,
|
| 211 |
+
"loss": 0.4142,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.68,
|
| 216 |
+
"grad_norm": 2.0792549392682624,
|
| 217 |
+
"learning_rate": 1e-05,
|
| 218 |
+
"loss": 0.3814,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.71,
|
| 223 |
+
"grad_norm": 2.4317303715104868,
|
| 224 |
+
"learning_rate": 1e-05,
|
| 225 |
+
"loss": 0.4042,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.73,
|
| 230 |
+
"grad_norm": 2.6323607639867443,
|
| 231 |
+
"learning_rate": 1e-05,
|
| 232 |
+
"loss": 0.3745,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.75,
|
| 237 |
+
"grad_norm": 2.2663973989673987,
|
| 238 |
+
"learning_rate": 1e-05,
|
| 239 |
+
"loss": 0.3781,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.77,
|
| 244 |
+
"grad_norm": 2.2288967526596193,
|
| 245 |
+
"learning_rate": 1e-05,
|
| 246 |
+
"loss": 0.3894,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.8,
|
| 251 |
+
"grad_norm": 2.2598649631650223,
|
| 252 |
+
"learning_rate": 1e-05,
|
| 253 |
+
"loss": 0.3734,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.82,
|
| 258 |
+
"grad_norm": 2.046710491098762,
|
| 259 |
+
"learning_rate": 1e-05,
|
| 260 |
+
"loss": 0.3436,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.84,
|
| 265 |
+
"grad_norm": 2.104660502205773,
|
| 266 |
+
"learning_rate": 1e-05,
|
| 267 |
+
"loss": 0.3636,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.87,
|
| 272 |
+
"grad_norm": 2.2194274529750335,
|
| 273 |
+
"learning_rate": 1e-05,
|
| 274 |
+
"loss": 0.3687,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.89,
|
| 279 |
+
"grad_norm": 2.380578087514372,
|
| 280 |
+
"learning_rate": 1e-05,
|
| 281 |
+
"loss": 0.3386,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.91,
|
| 286 |
+
"grad_norm": 2.1539059012744675,
|
| 287 |
+
"learning_rate": 1e-05,
|
| 288 |
+
"loss": 0.3663,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.93,
|
| 293 |
+
"grad_norm": 2.2975008924550884,
|
| 294 |
+
"learning_rate": 1e-05,
|
| 295 |
+
"loss": 0.3529,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.96,
|
| 300 |
+
"grad_norm": 2.0215009917813864,
|
| 301 |
+
"learning_rate": 1e-05,
|
| 302 |
+
"loss": 0.3255,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.98,
|
| 307 |
+
"grad_norm": 2.260279893370542,
|
| 308 |
+
"learning_rate": 1e-05,
|
| 309 |
+
"loss": 0.3496,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.0,
|
| 314 |
+
"grad_norm": 2.1998031778618596,
|
| 315 |
+
"learning_rate": 1e-05,
|
| 316 |
+
"loss": 0.3295,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.03,
|
| 321 |
+
"grad_norm": 1.9950078809811038,
|
| 322 |
+
"learning_rate": 1e-05,
|
| 323 |
+
"loss": 0.2919,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.05,
|
| 328 |
+
"grad_norm": 1.8802987071836594,
|
| 329 |
+
"learning_rate": 1e-05,
|
| 330 |
+
"loss": 0.2774,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.07,
|
| 335 |
+
"grad_norm": 2.330358996793001,
|
| 336 |
+
"learning_rate": 1e-05,
|
| 337 |
+
"loss": 0.2953,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.09,
|
| 342 |
+
"grad_norm": 2.187255642034782,
|
| 343 |
+
"learning_rate": 1e-05,
|
| 344 |
+
"loss": 0.2637,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.12,
|
| 349 |
+
"grad_norm": 2.246935471483204,
|
| 350 |
+
"learning_rate": 1e-05,
|
| 351 |
+
"loss": 0.2806,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.14,
|
| 356 |
+
"grad_norm": 2.129600005729623,
|
| 357 |
+
"learning_rate": 1e-05,
|
| 358 |
+
"loss": 0.2566,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.16,
|
| 363 |
+
"grad_norm": 2.1613544928929347,
|
| 364 |
+
"learning_rate": 1e-05,
|
| 365 |
+
"loss": 0.2703,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.19,
|
| 370 |
+
"grad_norm": 2.097651271845284,
|
| 371 |
+
"learning_rate": 1e-05,
|
| 372 |
+
"loss": 0.2578,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.21,
|
| 377 |
+
"grad_norm": 2.1698187501885404,
|
| 378 |
+
"learning_rate": 1e-05,
|
| 379 |
+
"loss": 0.266,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.23,
|
| 384 |
+
"grad_norm": 2.0534602028652733,
|
| 385 |
+
"learning_rate": 1e-05,
|
| 386 |
+
"loss": 0.2617,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.25,
|
| 391 |
+
"grad_norm": 2.280247094380378,
|
| 392 |
+
"learning_rate": 1e-05,
|
| 393 |
+
"loss": 0.2725,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.28,
|
| 398 |
+
"grad_norm": 2.282981855579156,
|
| 399 |
+
"learning_rate": 1e-05,
|
| 400 |
+
"loss": 0.2696,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.3,
|
| 405 |
+
"grad_norm": 2.1723199295603246,
|
| 406 |
+
"learning_rate": 1e-05,
|
| 407 |
+
"loss": 0.2801,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.32,
|
| 412 |
+
"grad_norm": 2.1013790638047833,
|
| 413 |
+
"learning_rate": 1e-05,
|
| 414 |
+
"loss": 0.2462,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.34,
|
| 419 |
+
"grad_norm": 2.1486192844939187,
|
| 420 |
+
"learning_rate": 1e-05,
|
| 421 |
+
"loss": 0.2634,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.37,
|
| 426 |
+
"grad_norm": 2.2011921615871874,
|
| 427 |
+
"learning_rate": 1e-05,
|
| 428 |
+
"loss": 0.2644,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.39,
|
| 433 |
+
"grad_norm": 2.0994036223733907,
|
| 434 |
+
"learning_rate": 1e-05,
|
| 435 |
+
"loss": 0.2366,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.41,
|
| 440 |
+
"grad_norm": 1.8682173502881247,
|
| 441 |
+
"learning_rate": 1e-05,
|
| 442 |
+
"loss": 0.2369,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.44,
|
| 447 |
+
"grad_norm": 1.9469094762833548,
|
| 448 |
+
"learning_rate": 1e-05,
|
| 449 |
+
"loss": 0.243,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.46,
|
| 454 |
+
"grad_norm": 2.198810958517713,
|
| 455 |
+
"learning_rate": 1e-05,
|
| 456 |
+
"loss": 0.2538,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.48,
|
| 461 |
+
"grad_norm": 1.9614171222986219,
|
| 462 |
+
"learning_rate": 1e-05,
|
| 463 |
+
"loss": 0.2209,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.5,
|
| 468 |
+
"grad_norm": 2.1909792689278924,
|
| 469 |
+
"learning_rate": 1e-05,
|
| 470 |
+
"loss": 0.2289,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.53,
|
| 475 |
+
"grad_norm": 2.219018585854138,
|
| 476 |
+
"learning_rate": 1e-05,
|
| 477 |
+
"loss": 0.2438,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.55,
|
| 482 |
+
"grad_norm": 2.095328922178155,
|
| 483 |
+
"learning_rate": 1e-05,
|
| 484 |
+
"loss": 0.2384,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.57,
|
| 489 |
+
"grad_norm": 2.1047747885459596,
|
| 490 |
+
"learning_rate": 1e-05,
|
| 491 |
+
"loss": 0.2511,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.6,
|
| 496 |
+
"grad_norm": 2.217218508465867,
|
| 497 |
+
"learning_rate": 1e-05,
|
| 498 |
+
"loss": 0.2326,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.62,
|
| 503 |
+
"grad_norm": 2.2940614424982364,
|
| 504 |
+
"learning_rate": 1e-05,
|
| 505 |
+
"loss": 0.2095,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.64,
|
| 510 |
+
"grad_norm": 2.0023231563553012,
|
| 511 |
+
"learning_rate": 1e-05,
|
| 512 |
+
"loss": 0.2178,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.66,
|
| 517 |
+
"grad_norm": 1.8803020902826912,
|
| 518 |
+
"learning_rate": 1e-05,
|
| 519 |
+
"loss": 0.2299,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.69,
|
| 524 |
+
"grad_norm": 1.9874036980658476,
|
| 525 |
+
"learning_rate": 1e-05,
|
| 526 |
+
"loss": 0.2085,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.71,
|
| 531 |
+
"grad_norm": 1.9993267804137187,
|
| 532 |
+
"learning_rate": 1e-05,
|
| 533 |
+
"loss": 0.2189,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.73,
|
| 538 |
+
"grad_norm": 2.075032021080106,
|
| 539 |
+
"learning_rate": 1e-05,
|
| 540 |
+
"loss": 0.2122,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.75,
|
| 545 |
+
"grad_norm": 1.9908316147368204,
|
| 546 |
+
"learning_rate": 1e-05,
|
| 547 |
+
"loss": 0.2169,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.78,
|
| 552 |
+
"grad_norm": 1.9737029328293805,
|
| 553 |
+
"learning_rate": 1e-05,
|
| 554 |
+
"loss": 0.1976,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.8,
|
| 559 |
+
"grad_norm": 2.069169880586868,
|
| 560 |
+
"learning_rate": 1e-05,
|
| 561 |
+
"loss": 0.2193,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.82,
|
| 566 |
+
"grad_norm": 1.9730042900890021,
|
| 567 |
+
"learning_rate": 1e-05,
|
| 568 |
+
"loss": 0.1959,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.85,
|
| 573 |
+
"grad_norm": 1.9487990391437768,
|
| 574 |
+
"learning_rate": 1e-05,
|
| 575 |
+
"loss": 0.2099,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.87,
|
| 580 |
+
"grad_norm": 2.2136709023857923,
|
| 581 |
+
"learning_rate": 1e-05,
|
| 582 |
+
"loss": 0.2064,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.89,
|
| 587 |
+
"grad_norm": 1.867453149915136,
|
| 588 |
+
"learning_rate": 1e-05,
|
| 589 |
+
"loss": 0.2148,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.91,
|
| 594 |
+
"grad_norm": 2.2341488312462374,
|
| 595 |
+
"learning_rate": 1e-05,
|
| 596 |
+
"loss": 0.2152,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.94,
|
| 601 |
+
"grad_norm": 1.917448307480571,
|
| 602 |
+
"learning_rate": 1e-05,
|
| 603 |
+
"loss": 0.2087,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.96,
|
| 608 |
+
"grad_norm": 2.1010824249510938,
|
| 609 |
+
"learning_rate": 1e-05,
|
| 610 |
+
"loss": 0.1989,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.98,
|
| 615 |
+
"grad_norm": 1.9948658919746771,
|
| 616 |
+
"learning_rate": 1e-05,
|
| 617 |
+
"loss": 0.184,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.01,
|
| 622 |
+
"grad_norm": 1.7700018111142861,
|
| 623 |
+
"learning_rate": 1e-05,
|
| 624 |
+
"loss": 0.1755,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.03,
|
| 629 |
+
"grad_norm": 1.7706928580089987,
|
| 630 |
+
"learning_rate": 1e-05,
|
| 631 |
+
"loss": 0.1524,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.05,
|
| 636 |
+
"grad_norm": 1.924950952533756,
|
| 637 |
+
"learning_rate": 1e-05,
|
| 638 |
+
"loss": 0.1639,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.07,
|
| 643 |
+
"grad_norm": 1.8520437499999318,
|
| 644 |
+
"learning_rate": 1e-05,
|
| 645 |
+
"loss": 0.1628,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 2.1,
|
| 650 |
+
"grad_norm": 1.9282660088754877,
|
| 651 |
+
"learning_rate": 1e-05,
|
| 652 |
+
"loss": 0.1738,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 2.12,
|
| 657 |
+
"grad_norm": 2.152060500211321,
|
| 658 |
+
"learning_rate": 1e-05,
|
| 659 |
+
"loss": 0.1533,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.14,
|
| 664 |
+
"grad_norm": 2.1072066102911204,
|
| 665 |
+
"learning_rate": 1e-05,
|
| 666 |
+
"loss": 0.1699,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.16,
|
| 671 |
+
"grad_norm": 1.9056526630981285,
|
| 672 |
+
"learning_rate": 1e-05,
|
| 673 |
+
"loss": 0.1542,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.19,
|
| 678 |
+
"grad_norm": 2.0541561878206815,
|
| 679 |
+
"learning_rate": 1e-05,
|
| 680 |
+
"loss": 0.1654,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.21,
|
| 685 |
+
"grad_norm": 1.9010874038699952,
|
| 686 |
+
"learning_rate": 1e-05,
|
| 687 |
+
"loss": 0.1284,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.23,
|
| 692 |
+
"grad_norm": 1.8180452474072457,
|
| 693 |
+
"learning_rate": 1e-05,
|
| 694 |
+
"loss": 0.1294,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 2.26,
|
| 699 |
+
"grad_norm": 1.9510507921351066,
|
| 700 |
+
"learning_rate": 1e-05,
|
| 701 |
+
"loss": 0.1555,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.28,
|
| 706 |
+
"grad_norm": 2.16385305667214,
|
| 707 |
+
"learning_rate": 1e-05,
|
| 708 |
+
"loss": 0.1568,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.3,
|
| 713 |
+
"grad_norm": 1.778578736374391,
|
| 714 |
+
"learning_rate": 1e-05,
|
| 715 |
+
"loss": 0.1493,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.32,
|
| 720 |
+
"grad_norm": 1.7020886640236346,
|
| 721 |
+
"learning_rate": 1e-05,
|
| 722 |
+
"loss": 0.1365,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.35,
|
| 727 |
+
"grad_norm": 1.9239197373726553,
|
| 728 |
+
"learning_rate": 1e-05,
|
| 729 |
+
"loss": 0.1529,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.37,
|
| 734 |
+
"grad_norm": 1.8300125885434804,
|
| 735 |
+
"learning_rate": 1e-05,
|
| 736 |
+
"loss": 0.1507,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.39,
|
| 741 |
+
"grad_norm": 1.7740533757030397,
|
| 742 |
+
"learning_rate": 1e-05,
|
| 743 |
+
"loss": 0.1404,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.42,
|
| 748 |
+
"grad_norm": 1.921607088125746,
|
| 749 |
+
"learning_rate": 1e-05,
|
| 750 |
+
"loss": 0.1403,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.44,
|
| 755 |
+
"grad_norm": 1.7244219031686432,
|
| 756 |
+
"learning_rate": 1e-05,
|
| 757 |
+
"loss": 0.1493,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.46,
|
| 762 |
+
"grad_norm": 1.840637334327782,
|
| 763 |
+
"learning_rate": 1e-05,
|
| 764 |
+
"loss": 0.1555,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.48,
|
| 769 |
+
"grad_norm": 1.7138834314865232,
|
| 770 |
+
"learning_rate": 1e-05,
|
| 771 |
+
"loss": 0.1507,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.51,
|
| 776 |
+
"grad_norm": 1.8869772327906467,
|
| 777 |
+
"learning_rate": 1e-05,
|
| 778 |
+
"loss": 0.1519,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.53,
|
| 783 |
+
"grad_norm": 1.722031710475424,
|
| 784 |
+
"learning_rate": 1e-05,
|
| 785 |
+
"loss": 0.1388,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.55,
|
| 790 |
+
"grad_norm": 1.8294129933303556,
|
| 791 |
+
"learning_rate": 1e-05,
|
| 792 |
+
"loss": 0.1429,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.58,
|
| 797 |
+
"grad_norm": 1.913994003850658,
|
| 798 |
+
"learning_rate": 1e-05,
|
| 799 |
+
"loss": 0.1391,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.6,
|
| 804 |
+
"grad_norm": 1.9673824351683147,
|
| 805 |
+
"learning_rate": 1e-05,
|
| 806 |
+
"loss": 0.1374,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.62,
|
| 811 |
+
"grad_norm": 1.9673399496447597,
|
| 812 |
+
"learning_rate": 1e-05,
|
| 813 |
+
"loss": 0.1458,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.64,
|
| 818 |
+
"grad_norm": 1.8027392959029291,
|
| 819 |
+
"learning_rate": 1e-05,
|
| 820 |
+
"loss": 0.1351,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.67,
|
| 825 |
+
"grad_norm": 1.667565922975832,
|
| 826 |
+
"learning_rate": 1e-05,
|
| 827 |
+
"loss": 0.1455,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.69,
|
| 832 |
+
"grad_norm": 1.8116940749067432,
|
| 833 |
+
"learning_rate": 1e-05,
|
| 834 |
+
"loss": 0.1502,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.71,
|
| 839 |
+
"grad_norm": 1.9922161139058774,
|
| 840 |
+
"learning_rate": 1e-05,
|
| 841 |
+
"loss": 0.1422,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.73,
|
| 846 |
+
"grad_norm": 2.0124481055349044,
|
| 847 |
+
"learning_rate": 1e-05,
|
| 848 |
+
"loss": 0.1373,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.76,
|
| 853 |
+
"grad_norm": 1.800014793589776,
|
| 854 |
+
"learning_rate": 1e-05,
|
| 855 |
+
"loss": 0.1342,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.78,
|
| 860 |
+
"grad_norm": 1.7532349977825892,
|
| 861 |
+
"learning_rate": 1e-05,
|
| 862 |
+
"loss": 0.1262,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.8,
|
| 867 |
+
"grad_norm": 1.7857219830778754,
|
| 868 |
+
"learning_rate": 1e-05,
|
| 869 |
+
"loss": 0.1375,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.83,
|
| 874 |
+
"grad_norm": 1.7910722257498661,
|
| 875 |
+
"learning_rate": 1e-05,
|
| 876 |
+
"loss": 0.148,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.85,
|
| 881 |
+
"grad_norm": 1.6958921797341877,
|
| 882 |
+
"learning_rate": 1e-05,
|
| 883 |
+
"loss": 0.1287,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.87,
|
| 888 |
+
"grad_norm": 1.9268954167687131,
|
| 889 |
+
"learning_rate": 1e-05,
|
| 890 |
+
"loss": 0.1334,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.89,
|
| 895 |
+
"grad_norm": 1.7925408289852067,
|
| 896 |
+
"learning_rate": 1e-05,
|
| 897 |
+
"loss": 0.1445,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.92,
|
| 902 |
+
"grad_norm": 1.9380903694792881,
|
| 903 |
+
"learning_rate": 1e-05,
|
| 904 |
+
"loss": 0.1478,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.94,
|
| 909 |
+
"grad_norm": 1.8550832947077076,
|
| 910 |
+
"learning_rate": 1e-05,
|
| 911 |
+
"loss": 0.1185,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.94,
|
| 916 |
+
"step": 129,
|
| 917 |
+
"total_flos": 56540883877888.0,
|
| 918 |
+
"train_loss": 0.30087860679441647,
|
| 919 |
+
"train_runtime": 2488.5787,
|
| 920 |
+
"train_samples_per_second": 6.613,
|
| 921 |
+
"train_steps_per_second": 0.052
|
| 922 |
+
}
|
| 923 |
+
],
|
| 924 |
+
"logging_steps": 1.0,
|
| 925 |
+
"max_steps": 129,
|
| 926 |
+
"num_input_tokens_seen": 0,
|
| 927 |
+
"num_train_epochs": 3,
|
| 928 |
+
"save_steps": 1.0,
|
| 929 |
+
"total_flos": 56540883877888.0,
|
| 930 |
+
"train_batch_size": 1,
|
| 931 |
+
"trial_name": null,
|
| 932 |
+
"trial_params": null
|
| 933 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ec0cf123f8636755d95d712de2b2741945a948cfb0736da83fa8c6f37034fb5
|
| 3 |
+
size 8056
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|