Model save
Browse files- README.md +4 -4
- adapter_config.json +5 -5
- adapter_model.safetensors +1 -1
- trainer_state.json +597 -44
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -34,11 +34,11 @@ More information needed
|
|
| 34 |
|
| 35 |
The following hyperparameters were used during training:
|
| 36 |
- learning_rate: 0.0001
|
| 37 |
-
- train_batch_size:
|
| 38 |
- eval_batch_size: 8
|
| 39 |
- seed: 42
|
| 40 |
- gradient_accumulation_steps: 16
|
| 41 |
-
- total_train_batch_size:
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
- lr_scheduler_warmup_steps: 2
|
|
@@ -50,7 +50,7 @@ The following hyperparameters were used during training:
|
|
| 50 |
|
| 51 |
### Framework versions
|
| 52 |
|
| 53 |
-
- PEFT 0.13.
|
| 54 |
- Transformers 4.45.2
|
| 55 |
- Pytorch 2.4.1
|
| 56 |
-
- Tokenizers 0.20.
|
|
|
|
| 34 |
|
| 35 |
The following hyperparameters were used during training:
|
| 36 |
- learning_rate: 0.0001
|
| 37 |
+
- train_batch_size: 6
|
| 38 |
- eval_batch_size: 8
|
| 39 |
- seed: 42
|
| 40 |
- gradient_accumulation_steps: 16
|
| 41 |
+
- total_train_batch_size: 96
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
- lr_scheduler_warmup_steps: 2
|
|
|
|
| 50 |
|
| 51 |
### Framework versions
|
| 52 |
|
| 53 |
+
- PEFT 0.13.2
|
| 54 |
- Transformers 4.45.2
|
| 55 |
- Pytorch 2.4.1
|
| 56 |
+
- Tokenizers 0.20.1
|
adapter_config.json
CHANGED
|
@@ -20,14 +20,14 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"gate_proj",
|
| 24 |
-
"v_proj",
|
| 25 |
"o_proj",
|
|
|
|
| 26 |
"up_proj",
|
| 27 |
"q_proj",
|
| 28 |
-
"
|
| 29 |
-
"
|
| 30 |
-
"
|
|
|
|
| 31 |
],
|
| 32 |
"task_type": "CAUSAL_LM",
|
| 33 |
"use_dora": false,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
|
|
|
|
|
|
| 23 |
"o_proj",
|
| 24 |
+
"down_proj",
|
| 25 |
"up_proj",
|
| 26 |
"q_proj",
|
| 27 |
+
"k_proj",
|
| 28 |
+
"gate_proj",
|
| 29 |
+
"v_proj",
|
| 30 |
+
"embed_tokens"
|
| 31 |
],
|
| 32 |
"task_type": "CAUSAL_LM",
|
| 33 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 555869984
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa56b2cb584e33c3b9e45c3a6a47be96d6dfe9276366a19a61f491823c915764
|
| 3 |
size 555869984
|
trainer_state.json
CHANGED
|
@@ -1,81 +1,634 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 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.
|
| 13 |
-
"grad_norm": 2.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 0.
|
| 16 |
"step": 250
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"epoch": 0.
|
| 20 |
-
"grad_norm":
|
| 21 |
-
"learning_rate":
|
| 22 |
-
"loss": 0.
|
| 23 |
"step": 500
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"epoch":
|
| 27 |
-
"grad_norm":
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 0.
|
| 30 |
"step": 750
|
| 31 |
},
|
| 32 |
{
|
| 33 |
-
"epoch":
|
| 34 |
-
"grad_norm":
|
| 35 |
-
"learning_rate":
|
| 36 |
-
"loss": 0.
|
| 37 |
"step": 1000
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"epoch":
|
| 41 |
-
"grad_norm":
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 0.
|
| 44 |
"step": 1250
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch":
|
| 48 |
-
"grad_norm":
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss": 0.
|
| 51 |
"step": 1500
|
| 52 |
},
|
| 53 |
{
|
| 54 |
-
"epoch":
|
| 55 |
-
"grad_norm": 2.
|
| 56 |
-
"learning_rate":
|
| 57 |
-
"loss": 0.
|
| 58 |
"step": 1750
|
| 59 |
},
|
| 60 |
{
|
| 61 |
-
"epoch":
|
| 62 |
-
"grad_norm":
|
| 63 |
-
"learning_rate":
|
| 64 |
-
"loss": 0.
|
| 65 |
"step": 2000
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"epoch":
|
| 69 |
-
"
|
| 70 |
-
"
|
| 71 |
-
"
|
| 72 |
-
"
|
| 73 |
-
|
| 74 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
}
|
| 76 |
],
|
| 77 |
"logging_steps": 250,
|
| 78 |
-
"max_steps":
|
| 79 |
"num_input_tokens_seen": 0,
|
| 80 |
"num_train_epochs": 4,
|
| 81 |
"save_steps": 500,
|
|
@@ -91,8 +644,8 @@
|
|
| 91 |
"attributes": {}
|
| 92 |
}
|
| 93 |
},
|
| 94 |
-
"total_flos":
|
| 95 |
-
"train_batch_size":
|
| 96 |
"trial_name": null,
|
| 97 |
"trial_params": null
|
| 98 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.9998179584034954,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 21972,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.045510399126200335,
|
| 13 |
+
"grad_norm": 2.0404157638549805,
|
| 14 |
+
"learning_rate": 9.887118798361402e-05,
|
| 15 |
+
"loss": 0.8444,
|
| 16 |
"step": 250
|
| 17 |
},
|
| 18 |
{
|
| 19 |
+
"epoch": 0.09102079825240067,
|
| 20 |
+
"grad_norm": 2.918970823287964,
|
| 21 |
+
"learning_rate": 9.773327264451525e-05,
|
| 22 |
+
"loss": 0.546,
|
| 23 |
"step": 500
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"epoch": 0.136531197378601,
|
| 27 |
+
"grad_norm": 2.195496082305908,
|
| 28 |
+
"learning_rate": 9.659535730541647e-05,
|
| 29 |
+
"loss": 0.5034,
|
| 30 |
"step": 750
|
| 31 |
},
|
| 32 |
{
|
| 33 |
+
"epoch": 0.18204159650480134,
|
| 34 |
+
"grad_norm": 2.3301496505737305,
|
| 35 |
+
"learning_rate": 9.54574419663177e-05,
|
| 36 |
+
"loss": 0.48,
|
| 37 |
"step": 1000
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"epoch": 0.22755199563100167,
|
| 41 |
+
"grad_norm": 2.254934549331665,
|
| 42 |
+
"learning_rate": 9.431952662721894e-05,
|
| 43 |
+
"loss": 0.462,
|
| 44 |
"step": 1250
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.273062394757202,
|
| 48 |
+
"grad_norm": 2.273568630218506,
|
| 49 |
+
"learning_rate": 9.318161128812017e-05,
|
| 50 |
+
"loss": 0.4531,
|
| 51 |
"step": 1500
|
| 52 |
},
|
| 53 |
{
|
| 54 |
+
"epoch": 0.31857279388340237,
|
| 55 |
+
"grad_norm": 2.2925281524658203,
|
| 56 |
+
"learning_rate": 9.204369594902141e-05,
|
| 57 |
+
"loss": 0.4442,
|
| 58 |
"step": 1750
|
| 59 |
},
|
| 60 |
{
|
| 61 |
+
"epoch": 0.3640831930096027,
|
| 62 |
+
"grad_norm": 1.9923436641693115,
|
| 63 |
+
"learning_rate": 9.090578060992263e-05,
|
| 64 |
+
"loss": 0.4346,
|
| 65 |
"step": 2000
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"epoch": 0.40959359213580304,
|
| 69 |
+
"grad_norm": 2.253584623336792,
|
| 70 |
+
"learning_rate": 8.976786527082386e-05,
|
| 71 |
+
"loss": 0.4286,
|
| 72 |
+
"step": 2250
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.45510399126200335,
|
| 76 |
+
"grad_norm": 2.1932711601257324,
|
| 77 |
+
"learning_rate": 8.862994993172509e-05,
|
| 78 |
+
"loss": 0.4201,
|
| 79 |
+
"step": 2500
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.5006143903882037,
|
| 83 |
+
"grad_norm": 2.224640130996704,
|
| 84 |
+
"learning_rate": 8.749203459262631e-05,
|
| 85 |
+
"loss": 0.4089,
|
| 86 |
+
"step": 2750
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.546124789514404,
|
| 90 |
+
"grad_norm": 2.7368321418762207,
|
| 91 |
+
"learning_rate": 8.635411925352754e-05,
|
| 92 |
+
"loss": 0.3989,
|
| 93 |
+
"step": 3000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.5916351886406044,
|
| 97 |
+
"grad_norm": 2.6221635341644287,
|
| 98 |
+
"learning_rate": 8.521620391442878e-05,
|
| 99 |
+
"loss": 0.3884,
|
| 100 |
+
"step": 3250
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.6371455877668047,
|
| 104 |
+
"grad_norm": 2.952885866165161,
|
| 105 |
+
"learning_rate": 8.407828857533e-05,
|
| 106 |
+
"loss": 0.3769,
|
| 107 |
+
"step": 3500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6826559868930051,
|
| 111 |
+
"grad_norm": 3.0590875148773193,
|
| 112 |
+
"learning_rate": 8.294037323623123e-05,
|
| 113 |
+
"loss": 0.3713,
|
| 114 |
+
"step": 3750
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7281663860192054,
|
| 118 |
+
"grad_norm": 2.773761034011841,
|
| 119 |
+
"learning_rate": 8.180245789713246e-05,
|
| 120 |
+
"loss": 0.3589,
|
| 121 |
+
"step": 4000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7736767851454057,
|
| 125 |
+
"grad_norm": 3.0175020694732666,
|
| 126 |
+
"learning_rate": 8.066454255803368e-05,
|
| 127 |
+
"loss": 0.3518,
|
| 128 |
+
"step": 4250
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.8191871842716061,
|
| 132 |
+
"grad_norm": 2.867734670639038,
|
| 133 |
+
"learning_rate": 7.952662721893491e-05,
|
| 134 |
+
"loss": 0.3433,
|
| 135 |
+
"step": 4500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.8646975833978064,
|
| 139 |
+
"grad_norm": 2.8594553470611572,
|
| 140 |
+
"learning_rate": 7.838871187983614e-05,
|
| 141 |
+
"loss": 0.3317,
|
| 142 |
+
"step": 4750
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.9102079825240067,
|
| 146 |
+
"grad_norm": 2.9394314289093018,
|
| 147 |
+
"learning_rate": 7.725079654073737e-05,
|
| 148 |
+
"loss": 0.3259,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.9557183816502071,
|
| 153 |
+
"grad_norm": 2.9321157932281494,
|
| 154 |
+
"learning_rate": 7.61128812016386e-05,
|
| 155 |
+
"loss": 0.3189,
|
| 156 |
+
"step": 5250
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.0012287807764073,
|
| 160 |
+
"grad_norm": 3.3222484588623047,
|
| 161 |
+
"learning_rate": 7.497496586253983e-05,
|
| 162 |
+
"loss": 0.3149,
|
| 163 |
+
"step": 5500
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.0467391799026078,
|
| 167 |
+
"grad_norm": 2.9881997108459473,
|
| 168 |
+
"learning_rate": 7.383705052344107e-05,
|
| 169 |
+
"loss": 0.3039,
|
| 170 |
+
"step": 5750
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.092249579028808,
|
| 174 |
+
"grad_norm": 3.13385272026062,
|
| 175 |
+
"learning_rate": 7.26991351843423e-05,
|
| 176 |
+
"loss": 0.2992,
|
| 177 |
+
"step": 6000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.1377599781550085,
|
| 181 |
+
"grad_norm": 3.6479313373565674,
|
| 182 |
+
"learning_rate": 7.156121984524352e-05,
|
| 183 |
+
"loss": 0.2918,
|
| 184 |
+
"step": 6250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.1832703772812088,
|
| 188 |
+
"grad_norm": 3.1848132610321045,
|
| 189 |
+
"learning_rate": 7.042330450614475e-05,
|
| 190 |
+
"loss": 0.2867,
|
| 191 |
+
"step": 6500
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 1.228780776407409,
|
| 195 |
+
"grad_norm": 3.0174155235290527,
|
| 196 |
+
"learning_rate": 6.928538916704597e-05,
|
| 197 |
+
"loss": 0.2841,
|
| 198 |
+
"step": 6750
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.2742911755336095,
|
| 202 |
+
"grad_norm": 3.2451624870300293,
|
| 203 |
+
"learning_rate": 6.814747382794721e-05,
|
| 204 |
+
"loss": 0.2795,
|
| 205 |
+
"step": 7000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.3198015746598097,
|
| 209 |
+
"grad_norm": 3.4281792640686035,
|
| 210 |
+
"learning_rate": 6.700955848884844e-05,
|
| 211 |
+
"loss": 0.2762,
|
| 212 |
+
"step": 7250
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.3653119737860102,
|
| 216 |
+
"grad_norm": 3.198448657989502,
|
| 217 |
+
"learning_rate": 6.587164314974966e-05,
|
| 218 |
+
"loss": 0.2713,
|
| 219 |
+
"step": 7500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.4108223729122105,
|
| 223 |
+
"grad_norm": 3.210265636444092,
|
| 224 |
+
"learning_rate": 6.473372781065089e-05,
|
| 225 |
+
"loss": 0.2695,
|
| 226 |
+
"step": 7750
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.4563327720384107,
|
| 230 |
+
"grad_norm": 2.9175002574920654,
|
| 231 |
+
"learning_rate": 6.359581247155212e-05,
|
| 232 |
+
"loss": 0.2645,
|
| 233 |
+
"step": 8000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.501843171164611,
|
| 237 |
+
"grad_norm": 2.8182640075683594,
|
| 238 |
+
"learning_rate": 6.245789713245334e-05,
|
| 239 |
+
"loss": 0.2646,
|
| 240 |
+
"step": 8250
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 1.5473535702908114,
|
| 244 |
+
"grad_norm": 2.8764665126800537,
|
| 245 |
+
"learning_rate": 6.131998179335457e-05,
|
| 246 |
+
"loss": 0.2608,
|
| 247 |
+
"step": 8500
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 1.592863969417012,
|
| 251 |
+
"grad_norm": 3.4266469478607178,
|
| 252 |
+
"learning_rate": 6.01820664542558e-05,
|
| 253 |
+
"loss": 0.2566,
|
| 254 |
+
"step": 8750
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.6383743685432122,
|
| 258 |
+
"grad_norm": 3.518105983734131,
|
| 259 |
+
"learning_rate": 5.904415111515703e-05,
|
| 260 |
+
"loss": 0.2526,
|
| 261 |
+
"step": 9000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.6838847676694124,
|
| 265 |
+
"grad_norm": 2.892153739929199,
|
| 266 |
+
"learning_rate": 5.790623577605826e-05,
|
| 267 |
+
"loss": 0.2495,
|
| 268 |
+
"step": 9250
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.7293951667956127,
|
| 272 |
+
"grad_norm": 2.9470598697662354,
|
| 273 |
+
"learning_rate": 5.676832043695949e-05,
|
| 274 |
+
"loss": 0.2478,
|
| 275 |
+
"step": 9500
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.7749055659218131,
|
| 279 |
+
"grad_norm": 2.7485554218292236,
|
| 280 |
+
"learning_rate": 5.563040509786073e-05,
|
| 281 |
+
"loss": 0.2475,
|
| 282 |
+
"step": 9750
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.8204159650480136,
|
| 286 |
+
"grad_norm": 2.946807622909546,
|
| 287 |
+
"learning_rate": 5.4492489758761954e-05,
|
| 288 |
+
"loss": 0.2434,
|
| 289 |
+
"step": 10000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.8659263641742139,
|
| 293 |
+
"grad_norm": 3.0199010372161865,
|
| 294 |
+
"learning_rate": 5.3354574419663187e-05,
|
| 295 |
+
"loss": 0.2411,
|
| 296 |
+
"step": 10250
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.9114367633004141,
|
| 300 |
+
"grad_norm": 2.9265036582946777,
|
| 301 |
+
"learning_rate": 5.221665908056441e-05,
|
| 302 |
+
"loss": 0.2388,
|
| 303 |
+
"step": 10500
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.9569471624266144,
|
| 307 |
+
"grad_norm": 3.0778729915618896,
|
| 308 |
+
"learning_rate": 5.107874374146564e-05,
|
| 309 |
+
"loss": 0.2384,
|
| 310 |
+
"step": 10750
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 2.0024575615528146,
|
| 314 |
+
"grad_norm": 2.8588147163391113,
|
| 315 |
+
"learning_rate": 4.9940828402366865e-05,
|
| 316 |
+
"loss": 0.2346,
|
| 317 |
+
"step": 11000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 2.0479679606790153,
|
| 321 |
+
"grad_norm": 2.6789040565490723,
|
| 322 |
+
"learning_rate": 4.88029130632681e-05,
|
| 323 |
+
"loss": 0.2281,
|
| 324 |
+
"step": 11250
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 2.0934783598052156,
|
| 328 |
+
"grad_norm": 2.937530279159546,
|
| 329 |
+
"learning_rate": 4.7664997724169324e-05,
|
| 330 |
+
"loss": 0.2275,
|
| 331 |
+
"step": 11500
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 2.138988758931416,
|
| 335 |
+
"grad_norm": 3.119990587234497,
|
| 336 |
+
"learning_rate": 4.652708238507055e-05,
|
| 337 |
+
"loss": 0.2232,
|
| 338 |
+
"step": 11750
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 2.184499158057616,
|
| 342 |
+
"grad_norm": 3.223001480102539,
|
| 343 |
+
"learning_rate": 4.5389167045971784e-05,
|
| 344 |
+
"loss": 0.2258,
|
| 345 |
+
"step": 12000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 2.2300095571838163,
|
| 349 |
+
"grad_norm": 3.1424496173858643,
|
| 350 |
+
"learning_rate": 4.425125170687301e-05,
|
| 351 |
+
"loss": 0.2208,
|
| 352 |
+
"step": 12250
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 2.275519956310017,
|
| 356 |
+
"grad_norm": 3.063213586807251,
|
| 357 |
+
"learning_rate": 4.3113336367774236e-05,
|
| 358 |
+
"loss": 0.2228,
|
| 359 |
+
"step": 12500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 2.3210303554362173,
|
| 363 |
+
"grad_norm": 2.8787999153137207,
|
| 364 |
+
"learning_rate": 4.197542102867547e-05,
|
| 365 |
+
"loss": 0.2214,
|
| 366 |
+
"step": 12750
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 2.3665407545624175,
|
| 370 |
+
"grad_norm": 2.8783469200134277,
|
| 371 |
+
"learning_rate": 4.08375056895767e-05,
|
| 372 |
+
"loss": 0.22,
|
| 373 |
+
"step": 13000
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 2.4120511536886178,
|
| 377 |
+
"grad_norm": 3.0098447799682617,
|
| 378 |
+
"learning_rate": 3.969959035047793e-05,
|
| 379 |
+
"loss": 0.2165,
|
| 380 |
+
"step": 13250
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 2.457561552814818,
|
| 384 |
+
"grad_norm": 3.0888121128082275,
|
| 385 |
+
"learning_rate": 3.8561675011379155e-05,
|
| 386 |
+
"loss": 0.2169,
|
| 387 |
+
"step": 13500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 2.5030719519410187,
|
| 391 |
+
"grad_norm": 2.7234771251678467,
|
| 392 |
+
"learning_rate": 3.742375967228038e-05,
|
| 393 |
+
"loss": 0.2144,
|
| 394 |
+
"step": 13750
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 2.548582351067219,
|
| 398 |
+
"grad_norm": 2.6636533737182617,
|
| 399 |
+
"learning_rate": 3.6285844333181614e-05,
|
| 400 |
+
"loss": 0.2116,
|
| 401 |
+
"step": 14000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 2.5940927501934192,
|
| 405 |
+
"grad_norm": 2.8893346786499023,
|
| 406 |
+
"learning_rate": 3.514792899408284e-05,
|
| 407 |
+
"loss": 0.2134,
|
| 408 |
+
"step": 14250
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 2.6396031493196195,
|
| 412 |
+
"grad_norm": 3.0065407752990723,
|
| 413 |
+
"learning_rate": 3.4010013654984066e-05,
|
| 414 |
+
"loss": 0.2125,
|
| 415 |
+
"step": 14500
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 2.6851135484458197,
|
| 419 |
+
"grad_norm": 3.033083200454712,
|
| 420 |
+
"learning_rate": 3.28720983158853e-05,
|
| 421 |
+
"loss": 0.2105,
|
| 422 |
+
"step": 14750
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 2.7306239475720204,
|
| 426 |
+
"grad_norm": 3.2030928134918213,
|
| 427 |
+
"learning_rate": 3.173418297678653e-05,
|
| 428 |
+
"loss": 0.2101,
|
| 429 |
+
"step": 15000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 2.7761343466982207,
|
| 433 |
+
"grad_norm": 2.713702917098999,
|
| 434 |
+
"learning_rate": 3.059626763768776e-05,
|
| 435 |
+
"loss": 0.2108,
|
| 436 |
+
"step": 15250
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.821644745824421,
|
| 440 |
+
"grad_norm": 3.2595670223236084,
|
| 441 |
+
"learning_rate": 2.9458352298588988e-05,
|
| 442 |
+
"loss": 0.2068,
|
| 443 |
+
"step": 15500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 2.867155144950621,
|
| 447 |
+
"grad_norm": 2.702853202819824,
|
| 448 |
+
"learning_rate": 2.8320436959490214e-05,
|
| 449 |
+
"loss": 0.2074,
|
| 450 |
+
"step": 15750
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 2.9126655440768214,
|
| 454 |
+
"grad_norm": 2.6552271842956543,
|
| 455 |
+
"learning_rate": 2.7182521620391444e-05,
|
| 456 |
+
"loss": 0.2057,
|
| 457 |
+
"step": 16000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 2.958175943203022,
|
| 461 |
+
"grad_norm": 2.71763014793396,
|
| 462 |
+
"learning_rate": 2.604460628129267e-05,
|
| 463 |
+
"loss": 0.2042,
|
| 464 |
+
"step": 16250
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 3.0036863423292224,
|
| 468 |
+
"grad_norm": 2.951164960861206,
|
| 469 |
+
"learning_rate": 2.4906690942193903e-05,
|
| 470 |
+
"loss": 0.2038,
|
| 471 |
+
"step": 16500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 3.0491967414554226,
|
| 475 |
+
"grad_norm": 2.5258820056915283,
|
| 476 |
+
"learning_rate": 2.376877560309513e-05,
|
| 477 |
+
"loss": 0.1984,
|
| 478 |
+
"step": 16750
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 3.094707140581623,
|
| 482 |
+
"grad_norm": 2.263176679611206,
|
| 483 |
+
"learning_rate": 2.263086026399636e-05,
|
| 484 |
+
"loss": 0.1978,
|
| 485 |
+
"step": 17000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 3.140217539707823,
|
| 489 |
+
"grad_norm": 2.466646194458008,
|
| 490 |
+
"learning_rate": 2.149294492489759e-05,
|
| 491 |
+
"loss": 0.1978,
|
| 492 |
+
"step": 17250
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 3.1857279388340234,
|
| 496 |
+
"grad_norm": 2.5945017337799072,
|
| 497 |
+
"learning_rate": 2.0355029585798818e-05,
|
| 498 |
+
"loss": 0.1979,
|
| 499 |
+
"step": 17500
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 3.231238337960224,
|
| 503 |
+
"grad_norm": 2.984452247619629,
|
| 504 |
+
"learning_rate": 1.9217114246700048e-05,
|
| 505 |
+
"loss": 0.1976,
|
| 506 |
+
"step": 17750
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 3.2767487370864243,
|
| 510 |
+
"grad_norm": 2.494358777999878,
|
| 511 |
+
"learning_rate": 1.8079198907601274e-05,
|
| 512 |
+
"loss": 0.1942,
|
| 513 |
+
"step": 18000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 3.3222591362126246,
|
| 517 |
+
"grad_norm": 2.498244047164917,
|
| 518 |
+
"learning_rate": 1.6941283568502504e-05,
|
| 519 |
+
"loss": 0.1954,
|
| 520 |
+
"step": 18250
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 3.367769535338825,
|
| 524 |
+
"grad_norm": 2.7161076068878174,
|
| 525 |
+
"learning_rate": 1.5803368229403733e-05,
|
| 526 |
+
"loss": 0.1945,
|
| 527 |
+
"step": 18500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 3.413279934465025,
|
| 531 |
+
"grad_norm": 2.229287624359131,
|
| 532 |
+
"learning_rate": 1.4665452890304963e-05,
|
| 533 |
+
"loss": 0.194,
|
| 534 |
+
"step": 18750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 3.458790333591226,
|
| 538 |
+
"grad_norm": 2.9624757766723633,
|
| 539 |
+
"learning_rate": 1.352753755120619e-05,
|
| 540 |
+
"loss": 0.1921,
|
| 541 |
+
"step": 19000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 3.504300732717426,
|
| 545 |
+
"grad_norm": 2.2083513736724854,
|
| 546 |
+
"learning_rate": 1.238962221210742e-05,
|
| 547 |
+
"loss": 0.1924,
|
| 548 |
+
"step": 19250
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 3.5498111318436263,
|
| 552 |
+
"grad_norm": 3.0686652660369873,
|
| 553 |
+
"learning_rate": 1.1251706873008648e-05,
|
| 554 |
+
"loss": 0.1907,
|
| 555 |
+
"step": 19500
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 3.5953215309698265,
|
| 559 |
+
"grad_norm": 2.542316198348999,
|
| 560 |
+
"learning_rate": 1.0113791533909878e-05,
|
| 561 |
+
"loss": 0.1918,
|
| 562 |
+
"step": 19750
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 3.640831930096027,
|
| 566 |
+
"grad_norm": 2.798839569091797,
|
| 567 |
+
"learning_rate": 8.975876194811106e-06,
|
| 568 |
+
"loss": 0.1903,
|
| 569 |
+
"step": 20000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 3.686342329222227,
|
| 573 |
+
"grad_norm": 2.9675209522247314,
|
| 574 |
+
"learning_rate": 7.837960855712335e-06,
|
| 575 |
+
"loss": 0.1898,
|
| 576 |
+
"step": 20250
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 3.7318527283484277,
|
| 580 |
+
"grad_norm": 3.0449047088623047,
|
| 581 |
+
"learning_rate": 6.700045516613563e-06,
|
| 582 |
+
"loss": 0.1899,
|
| 583 |
+
"step": 20500
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 3.777363127474628,
|
| 587 |
+
"grad_norm": 2.863095998764038,
|
| 588 |
+
"learning_rate": 5.562130177514793e-06,
|
| 589 |
+
"loss": 0.1887,
|
| 590 |
+
"step": 20750
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 3.8228735266008282,
|
| 594 |
+
"grad_norm": 2.4912309646606445,
|
| 595 |
+
"learning_rate": 4.4242148384160225e-06,
|
| 596 |
+
"loss": 0.1878,
|
| 597 |
+
"step": 21000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 3.8683839257270285,
|
| 601 |
+
"grad_norm": 2.7340869903564453,
|
| 602 |
+
"learning_rate": 3.286299499317251e-06,
|
| 603 |
+
"loss": 0.1889,
|
| 604 |
+
"step": 21250
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 3.9138943248532287,
|
| 608 |
+
"grad_norm": 2.823261022567749,
|
| 609 |
+
"learning_rate": 2.14838416021848e-06,
|
| 610 |
+
"loss": 0.1894,
|
| 611 |
+
"step": 21500
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 3.9594047239794294,
|
| 615 |
+
"grad_norm": 2.7040815353393555,
|
| 616 |
+
"learning_rate": 1.0104688211197086e-06,
|
| 617 |
+
"loss": 0.1874,
|
| 618 |
+
"step": 21750
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 3.9998179584034954,
|
| 622 |
+
"step": 21972,
|
| 623 |
+
"total_flos": 1.7842732433211018e+18,
|
| 624 |
+
"train_loss": 0.27372986671813043,
|
| 625 |
+
"train_runtime": 41504.9243,
|
| 626 |
+
"train_samples_per_second": 50.823,
|
| 627 |
+
"train_steps_per_second": 0.529
|
| 628 |
}
|
| 629 |
],
|
| 630 |
"logging_steps": 250,
|
| 631 |
+
"max_steps": 21972,
|
| 632 |
"num_input_tokens_seen": 0,
|
| 633 |
"num_train_epochs": 4,
|
| 634 |
"save_steps": 500,
|
|
|
|
| 644 |
"attributes": {}
|
| 645 |
}
|
| 646 |
},
|
| 647 |
+
"total_flos": 1.7842732433211018e+18,
|
| 648 |
+
"train_batch_size": 6,
|
| 649 |
"trial_name": null,
|
| 650 |
"trial_params": null
|
| 651 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d70ad05f2dd7bb2424b5a331d1ad81a4681bc57bd29bb412e98f4fbbc6036e3
|
| 3 |
size 5240
|