Commit
•
ba9cc5e
1
Parent(s):
e98e986
Model save
Browse files- README.md +8 -9
- all_results.json +6 -6
- config.json +1 -1
- model-00001-of-00003.safetensors +1 -1
- model-00002-of-00003.safetensors +1 -1
- model-00003-of-00003.safetensors +1 -1
- train_results.json +6 -6
- trainer_state.json +811 -18
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -2,15 +2,12 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: mistralai/Mistral-7B-v0.1
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
-
- trl
|
| 7 |
-
- sft
|
| 8 |
-
- generated_from_trainer
|
| 9 |
- trl
|
| 10 |
- sft
|
|
|
|
| 11 |
- generated_from_trainer
|
| 12 |
datasets:
|
| 13 |
-
-
|
| 14 |
model-index:
|
| 15 |
- name: mistral-safe-sft-full
|
| 16 |
results: []
|
|
@@ -21,9 +18,9 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 21 |
|
| 22 |
# mistral-safe-sft-full
|
| 23 |
|
| 24 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the
|
| 25 |
It achieves the following results on the evaluation set:
|
| 26 |
-
- Loss: 1.
|
| 27 |
|
| 28 |
## Model description
|
| 29 |
|
|
@@ -53,13 +50,15 @@ The following hyperparameters were used during training:
|
|
| 53 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 54 |
- lr_scheduler_type: cosine
|
| 55 |
- lr_scheduler_warmup_ratio: 0.1
|
| 56 |
-
- num_epochs:
|
| 57 |
|
| 58 |
### Training results
|
| 59 |
|
| 60 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 61 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 62 |
-
|
|
|
|
|
|
|
|
| 63 |
|
| 64 |
|
| 65 |
### Framework versions
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: mistralai/Mistral-7B-v0.1
|
| 4 |
tags:
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
+
- alignment-handbook
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
model-index:
|
| 12 |
- name: mistral-safe-sft-full
|
| 13 |
results: []
|
|
|
|
| 18 |
|
| 19 |
# mistral-safe-sft-full
|
| 20 |
|
| 21 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
|
| 22 |
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.7972
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
|
|
|
| 50 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
- lr_scheduler_type: cosine
|
| 52 |
- lr_scheduler_warmup_ratio: 0.1
|
| 53 |
+
- num_epochs: 4
|
| 54 |
|
| 55 |
### Training results
|
| 56 |
|
| 57 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 58 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 59 |
+
| 0.9697 | 2.0 | 370 | 1.2888 |
|
| 60 |
+
| 0.4114 | 3.0 | 555 | 1.4739 |
|
| 61 |
+
| 0.1361 | 4.0 | 740 | 1.7972 |
|
| 62 |
|
| 63 |
|
| 64 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,14 +1,14 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
"eval_loss": 1.2626361846923828,
|
| 4 |
"eval_runtime": 50.2072,
|
| 5 |
"eval_samples": 8340,
|
| 6 |
"eval_samples_per_second": 13.126,
|
| 7 |
"eval_steps_per_second": 0.418,
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
"train_samples": 75077,
|
| 12 |
-
"train_samples_per_second":
|
| 13 |
-
"train_steps_per_second": 0.
|
| 14 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
"eval_loss": 1.2626361846923828,
|
| 4 |
"eval_runtime": 50.2072,
|
| 5 |
"eval_samples": 8340,
|
| 6 |
"eval_samples_per_second": 13.126,
|
| 7 |
"eval_steps_per_second": 0.418,
|
| 8 |
+
"total_flos": 77470472601600.0,
|
| 9 |
+
"train_loss": 0.38161755220310106,
|
| 10 |
+
"train_runtime": 4944.5179,
|
| 11 |
"train_samples": 75077,
|
| 12 |
+
"train_samples_per_second": 4.784,
|
| 13 |
+
"train_steps_per_second": 0.15
|
| 14 |
}
|
config.json
CHANGED
|
@@ -22,6 +22,6 @@
|
|
| 22 |
"tie_word_embeddings": false,
|
| 23 |
"torch_dtype": "bfloat16",
|
| 24 |
"transformers_version": "4.43.3",
|
| 25 |
-
"use_cache":
|
| 26 |
"vocab_size": 32000
|
| 27 |
}
|
|
|
|
| 22 |
"tie_word_embeddings": false,
|
| 23 |
"torch_dtype": "bfloat16",
|
| 24 |
"transformers_version": "4.43.3",
|
| 25 |
+
"use_cache": false,
|
| 26 |
"vocab_size": 32000
|
| 27 |
}
|
model-00001-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4943162336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56346e353febfc862566359721f6f2426618fb66a58583bc8105791d50cfb0e6
|
| 3 |
size 4943162336
|
model-00002-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999819336
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f66cb79f32ece2a34d38226f847d166bf13e3f302e44eb5e0a3357f084839019
|
| 3 |
size 4999819336
|
model-00003-of-00003.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4540516344
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9aa3520d3011e838ebf851dc08edbc84caab425d5af2fb29cb8ee73fee69481
|
| 3 |
size 4540516344
|
train_results.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
"train_samples": 75077,
|
| 7 |
-
"train_samples_per_second":
|
| 8 |
-
"train_steps_per_second": 0.
|
| 9 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 77470472601600.0,
|
| 4 |
+
"train_loss": 0.38161755220310106,
|
| 5 |
+
"train_runtime": 4944.5179,
|
| 6 |
"train_samples": 75077,
|
| 7 |
+
"train_samples_per_second": 4.784,
|
| 8 |
+
"train_steps_per_second": 0.15
|
| 9 |
}
|
trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 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,
|
|
@@ -275,27 +275,820 @@
|
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"epoch": 1.
|
| 279 |
-
"
|
| 280 |
-
"
|
| 281 |
-
"
|
| 282 |
-
"
|
| 283 |
-
"step": 185
|
| 284 |
},
|
| 285 |
{
|
| 286 |
-
"epoch": 1.
|
| 287 |
-
"
|
| 288 |
-
"
|
| 289 |
-
"
|
| 290 |
-
"
|
| 291 |
-
|
| 292 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 293 |
}
|
| 294 |
],
|
| 295 |
"logging_steps": 5,
|
| 296 |
-
"max_steps":
|
| 297 |
"num_input_tokens_seen": 0,
|
| 298 |
-
"num_train_epochs":
|
| 299 |
"save_steps": 800,
|
| 300 |
"stateful_callbacks": {
|
| 301 |
"TrainerControl": {
|
|
@@ -309,7 +1102,7 @@
|
|
| 309 |
"attributes": {}
|
| 310 |
}
|
| 311 |
},
|
| 312 |
-
"total_flos":
|
| 313 |
"train_batch_size": 8,
|
| 314 |
"trial_name": null,
|
| 315 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 740,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 275 |
"step": 185
|
| 276 |
},
|
| 277 |
{
|
| 278 |
+
"epoch": 1.027027027027027,
|
| 279 |
+
"grad_norm": 5.410199324361583,
|
| 280 |
+
"learning_rate": 1.853992867931721e-05,
|
| 281 |
+
"loss": 0.944,
|
| 282 |
+
"step": 190
|
|
|
|
| 283 |
},
|
| 284 |
{
|
| 285 |
+
"epoch": 1.054054054054054,
|
| 286 |
+
"grad_norm": 5.152995409876598,
|
| 287 |
+
"learning_rate": 1.8414852973000503e-05,
|
| 288 |
+
"loss": 0.956,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 1.0810810810810811,
|
| 293 |
+
"grad_norm": 3.933917659184313,
|
| 294 |
+
"learning_rate": 1.8285096492438424e-05,
|
| 295 |
+
"loss": 0.9283,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 1.1081081081081081,
|
| 300 |
+
"grad_norm": 2.759086157792593,
|
| 301 |
+
"learning_rate": 1.8150731414862623e-05,
|
| 302 |
+
"loss": 0.9151,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.135135135135135,
|
| 307 |
+
"grad_norm": 2.910077750970637,
|
| 308 |
+
"learning_rate": 1.8011832481043577e-05,
|
| 309 |
+
"loss": 0.9413,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.1621621621621623,
|
| 314 |
+
"grad_norm": 3.1297209827761985,
|
| 315 |
+
"learning_rate": 1.78684769537159e-05,
|
| 316 |
+
"loss": 0.9243,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.1891891891891893,
|
| 321 |
+
"grad_norm": 3.406559269295354,
|
| 322 |
+
"learning_rate": 1.7720744574600865e-05,
|
| 323 |
+
"loss": 0.9474,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.2162162162162162,
|
| 328 |
+
"grad_norm": 3.2575862603317876,
|
| 329 |
+
"learning_rate": 1.756871752004992e-05,
|
| 330 |
+
"loss": 0.9461,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.2432432432432432,
|
| 335 |
+
"grad_norm": 3.2529118198639644,
|
| 336 |
+
"learning_rate": 1.7412480355334006e-05,
|
| 337 |
+
"loss": 0.9385,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.2702702702702702,
|
| 342 |
+
"grad_norm": 3.096374104973297,
|
| 343 |
+
"learning_rate": 1.7252119987603976e-05,
|
| 344 |
+
"loss": 0.9564,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.2972972972972974,
|
| 349 |
+
"grad_norm": 2.9780724324068695,
|
| 350 |
+
"learning_rate": 1.7087725617548385e-05,
|
| 351 |
+
"loss": 0.9456,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.3243243243243243,
|
| 356 |
+
"grad_norm": 2.887431987514818,
|
| 357 |
+
"learning_rate": 1.6919388689775463e-05,
|
| 358 |
+
"loss": 0.9608,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.3513513513513513,
|
| 363 |
+
"grad_norm": 3.420843173177096,
|
| 364 |
+
"learning_rate": 1.6747202841946928e-05,
|
| 365 |
+
"loss": 0.9817,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.3783783783783785,
|
| 370 |
+
"grad_norm": 2.817422226047505,
|
| 371 |
+
"learning_rate": 1.6571263852691887e-05,
|
| 372 |
+
"loss": 0.9794,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.4054054054054055,
|
| 377 |
+
"grad_norm": 2.8475793868122135,
|
| 378 |
+
"learning_rate": 1.639166958832985e-05,
|
| 379 |
+
"loss": 0.9249,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.4324324324324325,
|
| 384 |
+
"grad_norm": 3.084978828155213,
|
| 385 |
+
"learning_rate": 1.6208519948432438e-05,
|
| 386 |
+
"loss": 0.9694,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.4594594594594594,
|
| 391 |
+
"grad_norm": 2.9443084757542923,
|
| 392 |
+
"learning_rate": 1.6021916810254096e-05,
|
| 393 |
+
"loss": 0.9646,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.4864864864864864,
|
| 398 |
+
"grad_norm": 2.7978723842230555,
|
| 399 |
+
"learning_rate": 1.5831963972062734e-05,
|
| 400 |
+
"loss": 0.9614,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.5135135135135136,
|
| 405 |
+
"grad_norm": 2.8201773773365155,
|
| 406 |
+
"learning_rate": 1.5638767095401778e-05,
|
| 407 |
+
"loss": 0.9785,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.5405405405405406,
|
| 412 |
+
"grad_norm": 2.6511276981563907,
|
| 413 |
+
"learning_rate": 1.5442433646315792e-05,
|
| 414 |
+
"loss": 0.956,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.5675675675675675,
|
| 419 |
+
"grad_norm": 2.907386340027783,
|
| 420 |
+
"learning_rate": 1.5243072835572319e-05,
|
| 421 |
+
"loss": 0.9386,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.5945945945945947,
|
| 426 |
+
"grad_norm": 3.2379041081891455,
|
| 427 |
+
"learning_rate": 1.5040795557913246e-05,
|
| 428 |
+
"loss": 0.9891,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.6216216216216215,
|
| 433 |
+
"grad_norm": 2.659956186011908,
|
| 434 |
+
"learning_rate": 1.4835714330369445e-05,
|
| 435 |
+
"loss": 0.9806,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.6486486486486487,
|
| 440 |
+
"grad_norm": 2.794287463461555,
|
| 441 |
+
"learning_rate": 1.4627943229672992e-05,
|
| 442 |
+
"loss": 0.9856,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.6756756756756757,
|
| 447 |
+
"grad_norm": 2.9825237321465647,
|
| 448 |
+
"learning_rate": 1.4417597828801833e-05,
|
| 449 |
+
"loss": 1.011,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.7027027027027026,
|
| 454 |
+
"grad_norm": 2.6599834596055145,
|
| 455 |
+
"learning_rate": 1.4204795132692146e-05,
|
| 456 |
+
"loss": 0.9682,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.7297297297297298,
|
| 461 |
+
"grad_norm": 2.63763723653628,
|
| 462 |
+
"learning_rate": 1.3989653513154165e-05,
|
| 463 |
+
"loss": 0.9703,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.7567567567567568,
|
| 468 |
+
"grad_norm": 2.702492331093115,
|
| 469 |
+
"learning_rate": 1.37722926430277e-05,
|
| 470 |
+
"loss": 0.9695,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.7837837837837838,
|
| 475 |
+
"grad_norm": 2.6668927523262234,
|
| 476 |
+
"learning_rate": 1.3552833429613939e-05,
|
| 477 |
+
"loss": 0.9621,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.810810810810811,
|
| 482 |
+
"grad_norm": 2.830202295132874,
|
| 483 |
+
"learning_rate": 1.3331397947420578e-05,
|
| 484 |
+
"loss": 1.0006,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.8378378378378377,
|
| 489 |
+
"grad_norm": 2.960495369045161,
|
| 490 |
+
"learning_rate": 1.3108109370257714e-05,
|
| 491 |
+
"loss": 0.9658,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.864864864864865,
|
| 496 |
+
"grad_norm": 2.8089453201114005,
|
| 497 |
+
"learning_rate": 1.288309190272222e-05,
|
| 498 |
+
"loss": 0.978,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.8918918918918919,
|
| 503 |
+
"grad_norm": 2.690571777156308,
|
| 504 |
+
"learning_rate": 1.2656470711108763e-05,
|
| 505 |
+
"loss": 0.9694,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.9189189189189189,
|
| 510 |
+
"grad_norm": 3.072726161364189,
|
| 511 |
+
"learning_rate": 1.2428371853785872e-05,
|
| 512 |
+
"loss": 0.995,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.945945945945946,
|
| 517 |
+
"grad_norm": 2.650492212953216,
|
| 518 |
+
"learning_rate": 1.2198922211075779e-05,
|
| 519 |
+
"loss": 0.9633,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.972972972972973,
|
| 524 |
+
"grad_norm": 2.4600728300935524,
|
| 525 |
+
"learning_rate": 1.1968249414677055e-05,
|
| 526 |
+
"loss": 0.9518,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 2.0,
|
| 531 |
+
"grad_norm": 2.660921417467904,
|
| 532 |
+
"learning_rate": 1.1736481776669307e-05,
|
| 533 |
+
"loss": 0.9697,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.0,
|
| 538 |
+
"eval_loss": 1.2888436317443848,
|
| 539 |
+
"eval_runtime": 49.8535,
|
| 540 |
+
"eval_samples_per_second": 13.219,
|
| 541 |
+
"eval_steps_per_second": 0.421,
|
| 542 |
+
"step": 370
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 2.027027027027027,
|
| 546 |
+
"grad_norm": 4.650300715219466,
|
| 547 |
+
"learning_rate": 1.150374821813937e-05,
|
| 548 |
+
"loss": 0.4819,
|
| 549 |
+
"step": 375
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 2.054054054054054,
|
| 553 |
+
"grad_norm": 3.6035388443171055,
|
| 554 |
+
"learning_rate": 1.1270178197468788e-05,
|
| 555 |
+
"loss": 0.4496,
|
| 556 |
+
"step": 380
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 2.081081081081081,
|
| 560 |
+
"grad_norm": 3.2195791970064738,
|
| 561 |
+
"learning_rate": 1.1035901638322392e-05,
|
| 562 |
+
"loss": 0.4485,
|
| 563 |
+
"step": 385
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 2.108108108108108,
|
| 567 |
+
"grad_norm": 3.969479182342283,
|
| 568 |
+
"learning_rate": 1.080104885737807e-05,
|
| 569 |
+
"loss": 0.431,
|
| 570 |
+
"step": 390
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 2.135135135135135,
|
| 574 |
+
"grad_norm": 3.5220560829841108,
|
| 575 |
+
"learning_rate": 1.0565750491837925e-05,
|
| 576 |
+
"loss": 0.4358,
|
| 577 |
+
"step": 395
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 2.1621621621621623,
|
| 581 |
+
"grad_norm": 3.054414879545972,
|
| 582 |
+
"learning_rate": 1.0330137426761136e-05,
|
| 583 |
+
"loss": 0.4195,
|
| 584 |
+
"step": 400
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 2.189189189189189,
|
| 588 |
+
"grad_norm": 2.8407451960442844,
|
| 589 |
+
"learning_rate": 1.0094340722258969e-05,
|
| 590 |
+
"loss": 0.4139,
|
| 591 |
+
"step": 405
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 2.2162162162162162,
|
| 595 |
+
"grad_norm": 3.1380887590616835,
|
| 596 |
+
"learning_rate": 9.858491540592383e-06,
|
| 597 |
+
"loss": 0.4288,
|
| 598 |
+
"step": 410
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 2.2432432432432434,
|
| 602 |
+
"grad_norm": 3.0000811662471607,
|
| 603 |
+
"learning_rate": 9.622721073212831e-06,
|
| 604 |
+
"loss": 0.4307,
|
| 605 |
+
"step": 415
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 2.27027027027027,
|
| 609 |
+
"grad_norm": 2.9078632951242,
|
| 610 |
+
"learning_rate": 9.38716046778684e-06,
|
| 611 |
+
"loss": 0.4346,
|
| 612 |
+
"step": 420
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 2.2972972972972974,
|
| 616 |
+
"grad_norm": 3.0788575744315363,
|
| 617 |
+
"learning_rate": 9.151940755244912e-06,
|
| 618 |
+
"loss": 0.4354,
|
| 619 |
+
"step": 425
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 2.3243243243243246,
|
| 623 |
+
"grad_norm": 3.2452582293940764,
|
| 624 |
+
"learning_rate": 8.917192776895382e-06,
|
| 625 |
+
"loss": 0.43,
|
| 626 |
+
"step": 430
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.3513513513513513,
|
| 630 |
+
"grad_norm": 2.9959040185970345,
|
| 631 |
+
"learning_rate": 8.683047111643764e-06,
|
| 632 |
+
"loss": 0.4288,
|
| 633 |
+
"step": 435
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 2.3783783783783785,
|
| 637 |
+
"grad_norm": 2.9976789493802642,
|
| 638 |
+
"learning_rate": 8.449634003358022e-06,
|
| 639 |
+
"loss": 0.4154,
|
| 640 |
+
"step": 440
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.4054054054054053,
|
| 644 |
+
"grad_norm": 3.342720495579085,
|
| 645 |
+
"learning_rate": 8.217083288420241e-06,
|
| 646 |
+
"loss": 0.4184,
|
| 647 |
+
"step": 445
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 2.4324324324324325,
|
| 651 |
+
"grad_norm": 2.9954997310319817,
|
| 652 |
+
"learning_rate": 7.985524323504948e-06,
|
| 653 |
+
"loss": 0.4244,
|
| 654 |
+
"step": 450
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.4594594594594597,
|
| 658 |
+
"grad_norm": 3.1107476712409436,
|
| 659 |
+
"learning_rate": 7.755085913624274e-06,
|
| 660 |
+
"loss": 0.4201,
|
| 661 |
+
"step": 455
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.4864864864864864,
|
| 665 |
+
"grad_norm": 2.982814359917317,
|
| 666 |
+
"learning_rate": 7.525896240479977e-06,
|
| 667 |
+
"loss": 0.4223,
|
| 668 |
+
"step": 460
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.5135135135135136,
|
| 672 |
+
"grad_norm": 2.918536641642885,
|
| 673 |
+
"learning_rate": 7.29808279116218e-06,
|
| 674 |
+
"loss": 0.4355,
|
| 675 |
+
"step": 465
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.5405405405405403,
|
| 679 |
+
"grad_norm": 3.0794937831504914,
|
| 680 |
+
"learning_rate": 7.071772287234497e-06,
|
| 681 |
+
"loss": 0.4182,
|
| 682 |
+
"step": 470
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 2.5675675675675675,
|
| 686 |
+
"grad_norm": 3.350485105176484,
|
| 687 |
+
"learning_rate": 6.8470906142449764e-06,
|
| 688 |
+
"loss": 0.4173,
|
| 689 |
+
"step": 475
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 2.5945945945945947,
|
| 693 |
+
"grad_norm": 2.978358242778936,
|
| 694 |
+
"learning_rate": 6.624162751702077e-06,
|
| 695 |
+
"loss": 0.4257,
|
| 696 |
+
"step": 480
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.6216216216216215,
|
| 700 |
+
"grad_norm": 3.03990431149403,
|
| 701 |
+
"learning_rate": 6.403112703554643e-06,
|
| 702 |
+
"loss": 0.4198,
|
| 703 |
+
"step": 485
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 2.6486486486486487,
|
| 707 |
+
"grad_norm": 3.2537909886176797,
|
| 708 |
+
"learning_rate": 6.184063429214515e-06,
|
| 709 |
+
"loss": 0.4289,
|
| 710 |
+
"step": 490
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 2.6756756756756754,
|
| 714 |
+
"grad_norm": 2.9401645896134068,
|
| 715 |
+
"learning_rate": 5.967136775160188e-06,
|
| 716 |
+
"loss": 0.4229,
|
| 717 |
+
"step": 495
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.7027027027027026,
|
| 721 |
+
"grad_norm": 2.8796497097799385,
|
| 722 |
+
"learning_rate": 5.752453407159521e-06,
|
| 723 |
+
"loss": 0.4134,
|
| 724 |
+
"step": 500
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.72972972972973,
|
| 728 |
+
"grad_norm": 3.240398354047165,
|
| 729 |
+
"learning_rate": 5.5401327431492425e-06,
|
| 730 |
+
"loss": 0.4212,
|
| 731 |
+
"step": 505
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 2.756756756756757,
|
| 735 |
+
"grad_norm": 3.0875752772388556,
|
| 736 |
+
"learning_rate": 5.33029288680852e-06,
|
| 737 |
+
"loss": 0.4366,
|
| 738 |
+
"step": 510
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.7837837837837838,
|
| 742 |
+
"grad_norm": 2.950418599154064,
|
| 743 |
+
"learning_rate": 5.1230505618636575e-06,
|
| 744 |
+
"loss": 0.4132,
|
| 745 |
+
"step": 515
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 2.810810810810811,
|
| 749 |
+
"grad_norm": 2.9691405536481708,
|
| 750 |
+
"learning_rate": 4.918521047160309e-06,
|
| 751 |
+
"loss": 0.421,
|
| 752 |
+
"step": 520
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 2.8378378378378377,
|
| 756 |
+
"grad_norm": 3.288503082597523,
|
| 757 |
+
"learning_rate": 4.716818112539485e-06,
|
| 758 |
+
"loss": 0.4177,
|
| 759 |
+
"step": 525
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.864864864864865,
|
| 763 |
+
"grad_norm": 2.8513863896448304,
|
| 764 |
+
"learning_rate": 4.518053955552903e-06,
|
| 765 |
+
"loss": 0.4027,
|
| 766 |
+
"step": 530
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 2.891891891891892,
|
| 770 |
+
"grad_norm": 2.906451097506746,
|
| 771 |
+
"learning_rate": 4.322339139052922e-06,
|
| 772 |
+
"loss": 0.4104,
|
| 773 |
+
"step": 535
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 2.918918918918919,
|
| 777 |
+
"grad_norm": 2.8739958546058255,
|
| 778 |
+
"learning_rate": 4.1297825296918145e-06,
|
| 779 |
+
"loss": 0.4006,
|
| 780 |
+
"step": 540
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 2.945945945945946,
|
| 784 |
+
"grad_norm": 2.9696442817419113,
|
| 785 |
+
"learning_rate": 3.940491237364519e-06,
|
| 786 |
+
"loss": 0.4,
|
| 787 |
+
"step": 545
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 2.972972972972973,
|
| 791 |
+
"grad_norm": 3.0645448850809793,
|
| 792 |
+
"learning_rate": 3.754570555628613e-06,
|
| 793 |
+
"loss": 0.4043,
|
| 794 |
+
"step": 550
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 3.0,
|
| 798 |
+
"grad_norm": 2.9322335051873423,
|
| 799 |
+
"learning_rate": 3.5721239031346067e-06,
|
| 800 |
+
"loss": 0.4114,
|
| 801 |
+
"step": 555
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 3.0,
|
| 805 |
+
"eval_loss": 1.4739221334457397,
|
| 806 |
+
"eval_runtime": 47.6894,
|
| 807 |
+
"eval_samples_per_second": 13.819,
|
| 808 |
+
"eval_steps_per_second": 0.44,
|
| 809 |
+
"step": 555
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 3.027027027027027,
|
| 813 |
+
"grad_norm": 2.239642555101549,
|
| 814 |
+
"learning_rate": 3.3932527660991877e-06,
|
| 815 |
+
"loss": 0.1581,
|
| 816 |
+
"step": 560
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 3.054054054054054,
|
| 820 |
+
"grad_norm": 3.1082908400603464,
|
| 821 |
+
"learning_rate": 3.2180566418533365e-06,
|
| 822 |
+
"loss": 0.1515,
|
| 823 |
+
"step": 565
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 3.081081081081081,
|
| 827 |
+
"grad_norm": 2.152743576813358,
|
| 828 |
+
"learning_rate": 3.0466329834968234e-06,
|
| 829 |
+
"loss": 0.1438,
|
| 830 |
+
"step": 570
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 3.108108108108108,
|
| 834 |
+
"grad_norm": 2.232485171340327,
|
| 835 |
+
"learning_rate": 2.879077145689746e-06,
|
| 836 |
+
"loss": 0.1452,
|
| 837 |
+
"step": 575
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.135135135135135,
|
| 841 |
+
"grad_norm": 2.1315167934046464,
|
| 842 |
+
"learning_rate": 2.715482331611393e-06,
|
| 843 |
+
"loss": 0.1483,
|
| 844 |
+
"step": 580
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 3.1621621621621623,
|
| 848 |
+
"grad_norm": 1.9641583881749838,
|
| 849 |
+
"learning_rate": 2.5559395411158116e-06,
|
| 850 |
+
"loss": 0.1428,
|
| 851 |
+
"step": 585
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 3.189189189189189,
|
| 855 |
+
"grad_norm": 2.136285665522953,
|
| 856 |
+
"learning_rate": 2.4005375201130275e-06,
|
| 857 |
+
"loss": 0.1456,
|
| 858 |
+
"step": 590
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 3.2162162162162162,
|
| 862 |
+
"grad_norm": 2.3296203504449413,
|
| 863 |
+
"learning_rate": 2.249362711203985e-06,
|
| 864 |
+
"loss": 0.1451,
|
| 865 |
+
"step": 595
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 3.2432432432432434,
|
| 869 |
+
"grad_norm": 2.1953889356463105,
|
| 870 |
+
"learning_rate": 2.102499205596743e-06,
|
| 871 |
+
"loss": 0.1412,
|
| 872 |
+
"step": 600
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 3.27027027027027,
|
| 876 |
+
"grad_norm": 1.9846220715180103,
|
| 877 |
+
"learning_rate": 1.960028696330596e-06,
|
| 878 |
+
"loss": 0.139,
|
| 879 |
+
"step": 605
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 3.2972972972972974,
|
| 883 |
+
"grad_norm": 2.0955721946100345,
|
| 884 |
+
"learning_rate": 1.8220304328342253e-06,
|
| 885 |
+
"loss": 0.1379,
|
| 886 |
+
"step": 610
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 3.3243243243243246,
|
| 890 |
+
"grad_norm": 2.13188633069558,
|
| 891 |
+
"learning_rate": 1.688581176843066e-06,
|
| 892 |
+
"loss": 0.1401,
|
| 893 |
+
"step": 615
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 3.3513513513513513,
|
| 897 |
+
"grad_norm": 2.152817300804356,
|
| 898 |
+
"learning_rate": 1.5597551597004968e-06,
|
| 899 |
+
"loss": 0.1389,
|
| 900 |
+
"step": 620
|
| 901 |
+
},
|
| 902 |
+
{
|
| 903 |
+
"epoch": 3.3783783783783785,
|
| 904 |
+
"grad_norm": 2.0539946534038305,
|
| 905 |
+
"learning_rate": 1.4356240410665435e-06,
|
| 906 |
+
"loss": 0.141,
|
| 907 |
+
"step": 625
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 3.4054054054054053,
|
| 911 |
+
"grad_norm": 2.162669615931553,
|
| 912 |
+
"learning_rate": 1.3162568690570743e-06,
|
| 913 |
+
"loss": 0.1437,
|
| 914 |
+
"step": 630
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 3.4324324324324325,
|
| 918 |
+
"grad_norm": 2.153506447556338,
|
| 919 |
+
"learning_rate": 1.2017200418357077e-06,
|
| 920 |
+
"loss": 0.1417,
|
| 921 |
+
"step": 635
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 3.4594594594594597,
|
| 925 |
+
"grad_norm": 2.1701177657529964,
|
| 926 |
+
"learning_rate": 1.0920772706797166e-06,
|
| 927 |
+
"loss": 0.142,
|
| 928 |
+
"step": 640
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 3.4864864864864864,
|
| 932 |
+
"grad_norm": 2.0721670365744407,
|
| 933 |
+
"learning_rate": 9.873895445405523e-07,
|
| 934 |
+
"loss": 0.14,
|
| 935 |
+
"step": 645
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 3.5135135135135136,
|
| 939 |
+
"grad_norm": 2.0406246464567612,
|
| 940 |
+
"learning_rate": 8.87715096118642e-07,
|
| 941 |
+
"loss": 0.1398,
|
| 942 |
+
"step": 650
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 3.5405405405405403,
|
| 946 |
+
"grad_norm": 2.0428479721186035,
|
| 947 |
+
"learning_rate": 7.931093694713687e-07,
|
| 948 |
+
"loss": 0.1382,
|
| 949 |
+
"step": 655
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 3.5675675675675675,
|
| 953 |
+
"grad_norm": 2.234111473839529,
|
| 954 |
+
"learning_rate": 7.03624989172228e-07,
|
| 955 |
+
"loss": 0.1428,
|
| 956 |
+
"step": 660
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 3.5945945945945947,
|
| 960 |
+
"grad_norm": 1.9632293170603063,
|
| 961 |
+
"learning_rate": 6.193117310383412e-07,
|
| 962 |
+
"loss": 0.136,
|
| 963 |
+
"step": 665
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 3.6216216216216215,
|
| 967 |
+
"grad_norm": 2.0540773804658286,
|
| 968 |
+
"learning_rate": 5.402164944425758e-07,
|
| 969 |
+
"loss": 0.1355,
|
| 970 |
+
"step": 670
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 3.6486486486486487,
|
| 974 |
+
"grad_norm": 2.0789858869481144,
|
| 975 |
+
"learning_rate": 4.66383276225707e-07,
|
| 976 |
+
"loss": 0.1404,
|
| 977 |
+
"step": 675
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 3.6756756756756754,
|
| 981 |
+
"grad_norm": 2.2440192963067678,
|
| 982 |
+
"learning_rate": 3.97853146223105e-07,
|
| 983 |
+
"loss": 0.1384,
|
| 984 |
+
"step": 680
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 3.7027027027027026,
|
| 988 |
+
"grad_norm": 2.0732015708107703,
|
| 989 |
+
"learning_rate": 3.346642244195863e-07,
|
| 990 |
+
"loss": 0.1359,
|
| 991 |
+
"step": 685
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 3.72972972972973,
|
| 995 |
+
"grad_norm": 1.982186914698493,
|
| 996 |
+
"learning_rate": 2.7685165974510987e-07,
|
| 997 |
+
"loss": 0.1367,
|
| 998 |
+
"step": 690
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 3.756756756756757,
|
| 1002 |
+
"grad_norm": 2.0636323678600306,
|
| 1003 |
+
"learning_rate": 2.2444761052313857e-07,
|
| 1004 |
+
"loss": 0.136,
|
| 1005 |
+
"step": 695
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 3.7837837837837838,
|
| 1009 |
+
"grad_norm": 2.133975894121271,
|
| 1010 |
+
"learning_rate": 1.7748122658251877e-07,
|
| 1011 |
+
"loss": 0.1398,
|
| 1012 |
+
"step": 700
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 3.810810810810811,
|
| 1016 |
+
"grad_norm": 1.9445931920281623,
|
| 1017 |
+
"learning_rate": 1.3597863304285475e-07,
|
| 1018 |
+
"loss": 0.1384,
|
| 1019 |
+
"step": 705
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 3.8378378378378377,
|
| 1023 |
+
"grad_norm": 2.1247798692529414,
|
| 1024 |
+
"learning_rate": 9.996291578236228e-08,
|
| 1025 |
+
"loss": 0.1361,
|
| 1026 |
+
"step": 710
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 3.864864864864865,
|
| 1030 |
+
"grad_norm": 2.128261177290376,
|
| 1031 |
+
"learning_rate": 6.945410859632295e-08,
|
| 1032 |
+
"loss": 0.1326,
|
| 1033 |
+
"step": 715
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 3.891891891891892,
|
| 1037 |
+
"grad_norm": 2.059927006960735,
|
| 1038 |
+
"learning_rate": 4.44691820532539e-08,
|
| 1039 |
+
"loss": 0.1355,
|
| 1040 |
+
"step": 720
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 3.918918918918919,
|
| 1044 |
+
"grad_norm": 1.942567955434284,
|
| 1045 |
+
"learning_rate": 2.5022034055003363e-08,
|
| 1046 |
+
"loss": 0.1415,
|
| 1047 |
+
"step": 725
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 3.945945945945946,
|
| 1051 |
+
"grad_norm": 1.9857313689273417,
|
| 1052 |
+
"learning_rate": 1.1123482106021322e-08,
|
| 1053 |
+
"loss": 0.1356,
|
| 1054 |
+
"step": 730
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 3.972972972972973,
|
| 1058 |
+
"grad_norm": 2.0940341606759816,
|
| 1059 |
+
"learning_rate": 2.7812572961127825e-09,
|
| 1060 |
+
"loss": 0.1399,
|
| 1061 |
+
"step": 735
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 4.0,
|
| 1065 |
+
"grad_norm": 1.9231586161743661,
|
| 1066 |
+
"learning_rate": 0.0,
|
| 1067 |
+
"loss": 0.1361,
|
| 1068 |
+
"step": 740
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 4.0,
|
| 1072 |
+
"eval_loss": 1.7972235679626465,
|
| 1073 |
+
"eval_runtime": 46.1514,
|
| 1074 |
+
"eval_samples_per_second": 14.279,
|
| 1075 |
+
"eval_steps_per_second": 0.455,
|
| 1076 |
+
"step": 740
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 4.0,
|
| 1080 |
+
"step": 740,
|
| 1081 |
+
"total_flos": 77470472601600.0,
|
| 1082 |
+
"train_loss": 0.38161755220310106,
|
| 1083 |
+
"train_runtime": 4944.5179,
|
| 1084 |
+
"train_samples_per_second": 4.784,
|
| 1085 |
+
"train_steps_per_second": 0.15
|
| 1086 |
}
|
| 1087 |
],
|
| 1088 |
"logging_steps": 5,
|
| 1089 |
+
"max_steps": 740,
|
| 1090 |
"num_input_tokens_seen": 0,
|
| 1091 |
+
"num_train_epochs": 4,
|
| 1092 |
"save_steps": 800,
|
| 1093 |
"stateful_callbacks": {
|
| 1094 |
"TrainerControl": {
|
|
|
|
| 1102 |
"attributes": {}
|
| 1103 |
}
|
| 1104 |
},
|
| 1105 |
+
"total_flos": 77470472601600.0,
|
| 1106 |
"train_batch_size": 8,
|
| 1107 |
"trial_name": null,
|
| 1108 |
"trial_params": null
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 7032
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:777c461cc36b8e3230ad0d7aaa08c7521b9ebf955feac058b8fcada78fc62a2d
|
| 3 |
size 7032
|