End of training
Browse files- README.md +18 -6
- adapter_config.json +2 -2
- adapter_model.safetensors +1 -1
- loss_plot.png +0 -0
- trainer_state.json +1175 -29
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [deepseek-ai/deepseek-coder-1.3b-base](https://huggingface.co/deepseek-ai/deepseek-coder-1.3b-base) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -41,15 +41,27 @@ The following hyperparameters were used during training:
|
|
| 41 |
- seed: 42
|
| 42 |
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
-
-
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
-
| Training Loss | Epoch | Step
|
| 50 |
-
|
| 51 |
-
|
|
| 52 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
|
| 54 |
|
| 55 |
### Framework versions
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [deepseek-ai/deepseek-coder-1.3b-base](https://huggingface.co/deepseek-ai/deepseek-coder-1.3b-base) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.1481
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 41 |
- seed: 42
|
| 42 |
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
+
- num_epochs: 3
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
+
|:-------------:|:------:|:-----:|:---------------:|
|
| 51 |
+
| 0.3008 | 0.2000 | 5030 | 0.2892 |
|
| 52 |
+
| 0.2627 | 0.4001 | 10060 | 0.2511 |
|
| 53 |
+
| 0.2327 | 0.6001 | 15090 | 0.2334 |
|
| 54 |
+
| 0.2229 | 0.8001 | 20120 | 0.2116 |
|
| 55 |
+
| 0.2074 | 1.0002 | 25150 | 0.2014 |
|
| 56 |
+
| 0.1832 | 1.2002 | 30180 | 0.1921 |
|
| 57 |
+
| 0.1709 | 1.4002 | 35210 | 0.1860 |
|
| 58 |
+
| 0.1692 | 1.6003 | 40240 | 0.1758 |
|
| 59 |
+
| 0.1559 | 1.8003 | 45270 | 0.1679 |
|
| 60 |
+
| 0.1624 | 2.0003 | 50300 | 0.1645 |
|
| 61 |
+
| 0.1233 | 2.2003 | 55330 | 0.1600 |
|
| 62 |
+
| 0.1206 | 2.4004 | 60360 | 0.1539 |
|
| 63 |
+
| 0.1296 | 2.6004 | 65390 | 0.1516 |
|
| 64 |
+
| 0.1277 | 2.8004 | 70420 | 0.1481 |
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|
adapter_config.json
CHANGED
|
@@ -23,8 +23,8 @@
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
-
"
|
| 27 |
-
"
|
| 28 |
],
|
| 29 |
"task_type": "CAUSAL_LM",
|
| 30 |
"use_dora": false,
|
|
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
+
"v_proj",
|
| 27 |
+
"q_proj"
|
| 28 |
],
|
| 29 |
"task_type": "CAUSAL_LM",
|
| 30 |
"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 6304096
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a9be67420e23da7137c3deb8f4acab5bd0e8e89faa881700c0ffa81dac86c666
|
| 3 |
size 6304096
|
loss_plot.png
ADDED
|
trainer_state.json
CHANGED
|
@@ -1,43 +1,1189 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0
|
| 5 |
-
"eval_steps":
|
| 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 |
-
"
|
| 14 |
-
"
|
| 15 |
-
"
|
| 16 |
-
"
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
"
|
| 21 |
-
"
|
| 22 |
-
"
|
| 23 |
-
"
|
| 24 |
-
|
| 25 |
-
|
| 26 |
-
|
| 27 |
-
|
| 28 |
-
"
|
| 29 |
-
"
|
| 30 |
-
"
|
| 31 |
-
|
| 32 |
-
|
| 33 |
-
"
|
| 34 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
}
|
| 36 |
],
|
| 37 |
"logging_steps": 500,
|
| 38 |
-
"max_steps":
|
| 39 |
"num_input_tokens_seen": 0,
|
| 40 |
-
"num_train_epochs":
|
| 41 |
"save_steps": 500,
|
| 42 |
"stateful_callbacks": {
|
| 43 |
"TrainerControl": {
|
|
@@ -51,7 +1197,7 @@
|
|
| 51 |
"attributes": {}
|
| 52 |
}
|
| 53 |
},
|
| 54 |
-
"total_flos":
|
| 55 |
"train_batch_size": 2,
|
| 56 |
"trial_name": null,
|
| 57 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 5030,
|
| 6 |
+
"global_step": 75438,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
+
"epoch": 0.019883878151594685,
|
| 13 |
+
"grad_norm": 2.313262462615967,
|
| 14 |
+
"learning_rate": 0.00019867440812322703,
|
| 15 |
+
"loss": 0.5261,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03976775630318937,
|
| 20 |
+
"grad_norm": 2.2648062705993652,
|
| 21 |
+
"learning_rate": 0.00019734881624645405,
|
| 22 |
+
"loss": 0.4209,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05965163445478406,
|
| 27 |
+
"grad_norm": 2.1771059036254883,
|
| 28 |
+
"learning_rate": 0.0001960258755534346,
|
| 29 |
+
"loss": 0.3815,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07953551260637874,
|
| 34 |
+
"grad_norm": 1.579584002494812,
|
| 35 |
+
"learning_rate": 0.00019470028367666163,
|
| 36 |
+
"loss": 0.3557,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.09941939075797343,
|
| 41 |
+
"grad_norm": 0.9170318841934204,
|
| 42 |
+
"learning_rate": 0.00019337469179988865,
|
| 43 |
+
"loss": 0.3475,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.11930326890956812,
|
| 48 |
+
"grad_norm": 1.6301758289337158,
|
| 49 |
+
"learning_rate": 0.00019204909992311567,
|
| 50 |
+
"loss": 0.33,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1391871470611628,
|
| 55 |
+
"grad_norm": 1.6012169122695923,
|
| 56 |
+
"learning_rate": 0.0001907235080463427,
|
| 57 |
+
"loss": 0.3148,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.15907102521275748,
|
| 62 |
+
"grad_norm": 2.5444459915161133,
|
| 63 |
+
"learning_rate": 0.0001893979161695697,
|
| 64 |
+
"loss": 0.2942,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.17895490336435219,
|
| 69 |
+
"grad_norm": 1.427592396736145,
|
| 70 |
+
"learning_rate": 0.00018807232429279676,
|
| 71 |
+
"loss": 0.3047,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.19883878151594686,
|
| 76 |
+
"grad_norm": 2.3818204402923584,
|
| 77 |
+
"learning_rate": 0.00018674673241602375,
|
| 78 |
+
"loss": 0.3008,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.20003181420504254,
|
| 83 |
+
"eval_loss": 0.28922051191329956,
|
| 84 |
+
"eval_runtime": 32.072,
|
| 85 |
+
"eval_samples_per_second": 15.59,
|
| 86 |
+
"eval_steps_per_second": 7.795,
|
| 87 |
+
"step": 5030
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.21872265966754156,
|
| 91 |
+
"grad_norm": 1.6917686462402344,
|
| 92 |
+
"learning_rate": 0.0001854211405392508,
|
| 93 |
+
"loss": 0.2873,
|
| 94 |
+
"step": 5500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23860653781913624,
|
| 98 |
+
"grad_norm": 1.7427815198898315,
|
| 99 |
+
"learning_rate": 0.00018409819984623136,
|
| 100 |
+
"loss": 0.2893,
|
| 101 |
+
"step": 6000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.2584904159707309,
|
| 105 |
+
"grad_norm": 2.138218879699707,
|
| 106 |
+
"learning_rate": 0.00018277525915321192,
|
| 107 |
+
"loss": 0.2692,
|
| 108 |
+
"step": 6500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.2783742941223256,
|
| 112 |
+
"grad_norm": 1.8301661014556885,
|
| 113 |
+
"learning_rate": 0.00018144966727643894,
|
| 114 |
+
"loss": 0.2724,
|
| 115 |
+
"step": 7000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.2982581722739203,
|
| 119 |
+
"grad_norm": 1.288725733757019,
|
| 120 |
+
"learning_rate": 0.00018012407539966596,
|
| 121 |
+
"loss": 0.271,
|
| 122 |
+
"step": 7500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.31814205042551497,
|
| 126 |
+
"grad_norm": 1.311418056488037,
|
| 127 |
+
"learning_rate": 0.00017880113470664652,
|
| 128 |
+
"loss": 0.2701,
|
| 129 |
+
"step": 8000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.33802592857710967,
|
| 133 |
+
"grad_norm": 1.3854721784591675,
|
| 134 |
+
"learning_rate": 0.00017747554282987354,
|
| 135 |
+
"loss": 0.2598,
|
| 136 |
+
"step": 8500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.35790980672870437,
|
| 140 |
+
"grad_norm": 1.3663103580474854,
|
| 141 |
+
"learning_rate": 0.00017614995095310056,
|
| 142 |
+
"loss": 0.257,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3777936848802991,
|
| 147 |
+
"grad_norm": 1.9446492195129395,
|
| 148 |
+
"learning_rate": 0.00017482435907632758,
|
| 149 |
+
"loss": 0.2594,
|
| 150 |
+
"step": 9500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3976775630318937,
|
| 154 |
+
"grad_norm": 3.062659978866577,
|
| 155 |
+
"learning_rate": 0.0001734987671995546,
|
| 156 |
+
"loss": 0.2627,
|
| 157 |
+
"step": 10000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4000636284100851,
|
| 161 |
+
"eval_loss": 0.25110873579978943,
|
| 162 |
+
"eval_runtime": 31.6493,
|
| 163 |
+
"eval_samples_per_second": 15.798,
|
| 164 |
+
"eval_steps_per_second": 7.899,
|
| 165 |
+
"step": 10060
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.4175614411834884,
|
| 169 |
+
"grad_norm": 1.6593842506408691,
|
| 170 |
+
"learning_rate": 0.00017217317532278162,
|
| 171 |
+
"loss": 0.2495,
|
| 172 |
+
"step": 10500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.4374453193350831,
|
| 176 |
+
"grad_norm": 3.293686628341675,
|
| 177 |
+
"learning_rate": 0.00017084758344600864,
|
| 178 |
+
"loss": 0.2546,
|
| 179 |
+
"step": 11000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.4573291974866778,
|
| 183 |
+
"grad_norm": 1.244734764099121,
|
| 184 |
+
"learning_rate": 0.00016952199156923566,
|
| 185 |
+
"loss": 0.2537,
|
| 186 |
+
"step": 11500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.4772130756382725,
|
| 190 |
+
"grad_norm": 2.361588954925537,
|
| 191 |
+
"learning_rate": 0.00016819905087621622,
|
| 192 |
+
"loss": 0.2551,
|
| 193 |
+
"step": 12000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.4970969537898672,
|
| 197 |
+
"grad_norm": 0.7586632966995239,
|
| 198 |
+
"learning_rate": 0.00016687345899944327,
|
| 199 |
+
"loss": 0.2454,
|
| 200 |
+
"step": 12500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.5169808319414618,
|
| 204 |
+
"grad_norm": 2.943408250808716,
|
| 205 |
+
"learning_rate": 0.00016554786712267026,
|
| 206 |
+
"loss": 0.2445,
|
| 207 |
+
"step": 13000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.5368647100930566,
|
| 211 |
+
"grad_norm": 1.9744367599487305,
|
| 212 |
+
"learning_rate": 0.0001642222752458973,
|
| 213 |
+
"loss": 0.2373,
|
| 214 |
+
"step": 13500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.5567485882446512,
|
| 218 |
+
"grad_norm": 1.9832626581192017,
|
| 219 |
+
"learning_rate": 0.0001628966833691243,
|
| 220 |
+
"loss": 0.2417,
|
| 221 |
+
"step": 14000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.5766324663962459,
|
| 225 |
+
"grad_norm": 2.04445481300354,
|
| 226 |
+
"learning_rate": 0.00016157109149235135,
|
| 227 |
+
"loss": 0.2354,
|
| 228 |
+
"step": 14500
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.5965163445478406,
|
| 232 |
+
"grad_norm": 1.9727308750152588,
|
| 233 |
+
"learning_rate": 0.0001602481507993319,
|
| 234 |
+
"loss": 0.2327,
|
| 235 |
+
"step": 15000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.6000954426151277,
|
| 239 |
+
"eval_loss": 0.23340041935443878,
|
| 240 |
+
"eval_runtime": 32.528,
|
| 241 |
+
"eval_samples_per_second": 15.371,
|
| 242 |
+
"eval_steps_per_second": 7.686,
|
| 243 |
+
"step": 15090
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.6164002226994353,
|
| 247 |
+
"grad_norm": 1.3546355962753296,
|
| 248 |
+
"learning_rate": 0.00015892255892255893,
|
| 249 |
+
"loss": 0.2286,
|
| 250 |
+
"step": 15500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.6362841008510299,
|
| 254 |
+
"grad_norm": 1.3803051710128784,
|
| 255 |
+
"learning_rate": 0.0001575996182295395,
|
| 256 |
+
"loss": 0.2314,
|
| 257 |
+
"step": 16000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.6561679790026247,
|
| 261 |
+
"grad_norm": 1.1035007238388062,
|
| 262 |
+
"learning_rate": 0.00015627402635276651,
|
| 263 |
+
"loss": 0.2344,
|
| 264 |
+
"step": 16500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.6760518571542193,
|
| 268 |
+
"grad_norm": 1.2522505521774292,
|
| 269 |
+
"learning_rate": 0.00015494843447599353,
|
| 270 |
+
"loss": 0.2304,
|
| 271 |
+
"step": 17000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.6959357353058141,
|
| 275 |
+
"grad_norm": 1.4340494871139526,
|
| 276 |
+
"learning_rate": 0.00015362284259922055,
|
| 277 |
+
"loss": 0.226,
|
| 278 |
+
"step": 17500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.7158196134574087,
|
| 282 |
+
"grad_norm": 1.3514714241027832,
|
| 283 |
+
"learning_rate": 0.00015229725072244757,
|
| 284 |
+
"loss": 0.2282,
|
| 285 |
+
"step": 18000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.7357034916090034,
|
| 289 |
+
"grad_norm": 2.4679174423217773,
|
| 290 |
+
"learning_rate": 0.0001509716588456746,
|
| 291 |
+
"loss": 0.2199,
|
| 292 |
+
"step": 18500
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.7555873697605981,
|
| 296 |
+
"grad_norm": 2.4193334579467773,
|
| 297 |
+
"learning_rate": 0.0001496460669689016,
|
| 298 |
+
"loss": 0.2213,
|
| 299 |
+
"step": 19000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.7754712479121928,
|
| 303 |
+
"grad_norm": 1.094509482383728,
|
| 304 |
+
"learning_rate": 0.00014832047509212866,
|
| 305 |
+
"loss": 0.2163,
|
| 306 |
+
"step": 19500
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.7953551260637874,
|
| 310 |
+
"grad_norm": 1.7273403406143188,
|
| 311 |
+
"learning_rate": 0.00014699753439910922,
|
| 312 |
+
"loss": 0.2229,
|
| 313 |
+
"step": 20000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.8001272568201702,
|
| 317 |
+
"eval_loss": 0.21162502467632294,
|
| 318 |
+
"eval_runtime": 34.1481,
|
| 319 |
+
"eval_samples_per_second": 14.642,
|
| 320 |
+
"eval_steps_per_second": 7.321,
|
| 321 |
+
"step": 20120
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.8152390042153822,
|
| 325 |
+
"grad_norm": 2.999072313308716,
|
| 326 |
+
"learning_rate": 0.00014567194252233621,
|
| 327 |
+
"loss": 0.2203,
|
| 328 |
+
"step": 20500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.8351228823669768,
|
| 332 |
+
"grad_norm": 1.7195533514022827,
|
| 333 |
+
"learning_rate": 0.00014434635064556326,
|
| 334 |
+
"loss": 0.2121,
|
| 335 |
+
"step": 21000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.8550067605185715,
|
| 339 |
+
"grad_norm": 0.6770505905151367,
|
| 340 |
+
"learning_rate": 0.00014302075876879025,
|
| 341 |
+
"loss": 0.2045,
|
| 342 |
+
"step": 21500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.8748906386701663,
|
| 346 |
+
"grad_norm": 1.6488806009292603,
|
| 347 |
+
"learning_rate": 0.0001416951668920173,
|
| 348 |
+
"loss": 0.2062,
|
| 349 |
+
"step": 22000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.8947745168217609,
|
| 353 |
+
"grad_norm": 1.1503865718841553,
|
| 354 |
+
"learning_rate": 0.00014037222619899786,
|
| 355 |
+
"loss": 0.2134,
|
| 356 |
+
"step": 22500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.9146583949733555,
|
| 360 |
+
"grad_norm": 1.1807224750518799,
|
| 361 |
+
"learning_rate": 0.00013904663432222488,
|
| 362 |
+
"loss": 0.2131,
|
| 363 |
+
"step": 23000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.9345422731249503,
|
| 367 |
+
"grad_norm": 0.2355421781539917,
|
| 368 |
+
"learning_rate": 0.0001377210424454519,
|
| 369 |
+
"loss": 0.2185,
|
| 370 |
+
"step": 23500
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.954426151276545,
|
| 374 |
+
"grad_norm": 1.5836342573165894,
|
| 375 |
+
"learning_rate": 0.00013639545056867892,
|
| 376 |
+
"loss": 0.2122,
|
| 377 |
+
"step": 24000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.9743100294281397,
|
| 381 |
+
"grad_norm": 1.7692372798919678,
|
| 382 |
+
"learning_rate": 0.00013506985869190594,
|
| 383 |
+
"loss": 0.1968,
|
| 384 |
+
"step": 24500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.9941939075797344,
|
| 388 |
+
"grad_norm": 1.8389042615890503,
|
| 389 |
+
"learning_rate": 0.0001337469179988865,
|
| 390 |
+
"loss": 0.2074,
|
| 391 |
+
"step": 25000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.0001590710252128,
|
| 395 |
+
"eval_loss": 0.2013542503118515,
|
| 396 |
+
"eval_runtime": 32.8369,
|
| 397 |
+
"eval_samples_per_second": 15.227,
|
| 398 |
+
"eval_steps_per_second": 7.613,
|
| 399 |
+
"step": 25150
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.014077785731329,
|
| 403 |
+
"grad_norm": 0.7946391701698303,
|
| 404 |
+
"learning_rate": 0.00013242132612211352,
|
| 405 |
+
"loss": 0.1843,
|
| 406 |
+
"step": 25500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.0339616638829237,
|
| 410 |
+
"grad_norm": 1.3439126014709473,
|
| 411 |
+
"learning_rate": 0.00013109573424534054,
|
| 412 |
+
"loss": 0.1866,
|
| 413 |
+
"step": 26000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.0538455420345185,
|
| 417 |
+
"grad_norm": 2.3524932861328125,
|
| 418 |
+
"learning_rate": 0.00012977014236856756,
|
| 419 |
+
"loss": 0.1872,
|
| 420 |
+
"step": 26500
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.0737294201861132,
|
| 424 |
+
"grad_norm": 1.4595142602920532,
|
| 425 |
+
"learning_rate": 0.0001284445504917946,
|
| 426 |
+
"loss": 0.1857,
|
| 427 |
+
"step": 27000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.0936132983377078,
|
| 431 |
+
"grad_norm": 2.1640143394470215,
|
| 432 |
+
"learning_rate": 0.0001271189586150216,
|
| 433 |
+
"loss": 0.1926,
|
| 434 |
+
"step": 27500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.1134971764893025,
|
| 438 |
+
"grad_norm": 1.9392280578613281,
|
| 439 |
+
"learning_rate": 0.00012579336673824865,
|
| 440 |
+
"loss": 0.1891,
|
| 441 |
+
"step": 28000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.1333810546408971,
|
| 445 |
+
"grad_norm": 2.7630889415740967,
|
| 446 |
+
"learning_rate": 0.00012446777486147564,
|
| 447 |
+
"loss": 0.1878,
|
| 448 |
+
"step": 28500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.1532649327924918,
|
| 452 |
+
"grad_norm": 2.0248076915740967,
|
| 453 |
+
"learning_rate": 0.0001231448341684562,
|
| 454 |
+
"loss": 0.1866,
|
| 455 |
+
"step": 29000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.1731488109440866,
|
| 459 |
+
"grad_norm": 2.3321919441223145,
|
| 460 |
+
"learning_rate": 0.00012181924229168324,
|
| 461 |
+
"loss": 0.1876,
|
| 462 |
+
"step": 29500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.1930326890956813,
|
| 466 |
+
"grad_norm": 1.0356630086898804,
|
| 467 |
+
"learning_rate": 0.00012049365041491026,
|
| 468 |
+
"loss": 0.1832,
|
| 469 |
+
"step": 30000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.2001908852302554,
|
| 473 |
+
"eval_loss": 0.19214199483394623,
|
| 474 |
+
"eval_runtime": 32.7215,
|
| 475 |
+
"eval_samples_per_second": 15.28,
|
| 476 |
+
"eval_steps_per_second": 7.64,
|
| 477 |
+
"step": 30180
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.212916567247276,
|
| 481 |
+
"grad_norm": 1.0846548080444336,
|
| 482 |
+
"learning_rate": 0.00011916805853813728,
|
| 483 |
+
"loss": 0.1812,
|
| 484 |
+
"step": 30500
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 1.2328004453988706,
|
| 488 |
+
"grad_norm": 0.9677469730377197,
|
| 489 |
+
"learning_rate": 0.00011784511784511784,
|
| 490 |
+
"loss": 0.1812,
|
| 491 |
+
"step": 31000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.2526843235504652,
|
| 495 |
+
"grad_norm": 1.260237216949463,
|
| 496 |
+
"learning_rate": 0.00011651952596834487,
|
| 497 |
+
"loss": 0.1818,
|
| 498 |
+
"step": 31500
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.27256820170206,
|
| 502 |
+
"grad_norm": 1.1330456733703613,
|
| 503 |
+
"learning_rate": 0.00011519658527532544,
|
| 504 |
+
"loss": 0.1822,
|
| 505 |
+
"step": 32000
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.2924520798536547,
|
| 509 |
+
"grad_norm": 2.064711570739746,
|
| 510 |
+
"learning_rate": 0.00011387099339855246,
|
| 511 |
+
"loss": 0.1847,
|
| 512 |
+
"step": 32500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.3123359580052494,
|
| 516 |
+
"grad_norm": 1.2208553552627563,
|
| 517 |
+
"learning_rate": 0.00011254540152177948,
|
| 518 |
+
"loss": 0.1795,
|
| 519 |
+
"step": 33000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.332219836156844,
|
| 523 |
+
"grad_norm": 1.7120323181152344,
|
| 524 |
+
"learning_rate": 0.0001112198096450065,
|
| 525 |
+
"loss": 0.1768,
|
| 526 |
+
"step": 33500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.3521037143084387,
|
| 530 |
+
"grad_norm": 1.3297480344772339,
|
| 531 |
+
"learning_rate": 0.00010989686895198707,
|
| 532 |
+
"loss": 0.184,
|
| 533 |
+
"step": 34000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.3719875924600333,
|
| 537 |
+
"grad_norm": 2.436786651611328,
|
| 538 |
+
"learning_rate": 0.00010857127707521408,
|
| 539 |
+
"loss": 0.1837,
|
| 540 |
+
"step": 34500
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 1.391871470611628,
|
| 544 |
+
"grad_norm": 1.982055902481079,
|
| 545 |
+
"learning_rate": 0.00010724568519844111,
|
| 546 |
+
"loss": 0.1709,
|
| 547 |
+
"step": 35000
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.400222699435298,
|
| 551 |
+
"eval_loss": 0.18601036071777344,
|
| 552 |
+
"eval_runtime": 33.3133,
|
| 553 |
+
"eval_samples_per_second": 15.009,
|
| 554 |
+
"eval_steps_per_second": 7.505,
|
| 555 |
+
"step": 35210
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.4117553487632228,
|
| 559 |
+
"grad_norm": 1.3936830759048462,
|
| 560 |
+
"learning_rate": 0.00010592009332166812,
|
| 561 |
+
"loss": 0.1792,
|
| 562 |
+
"step": 35500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.4316392269148175,
|
| 566 |
+
"grad_norm": 2.4283194541931152,
|
| 567 |
+
"learning_rate": 0.00010459715262864868,
|
| 568 |
+
"loss": 0.1837,
|
| 569 |
+
"step": 36000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.4515231050664121,
|
| 573 |
+
"grad_norm": 1.2691034078598022,
|
| 574 |
+
"learning_rate": 0.00010327156075187571,
|
| 575 |
+
"loss": 0.1734,
|
| 576 |
+
"step": 36500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.4714069832180068,
|
| 580 |
+
"grad_norm": 0.4714362919330597,
|
| 581 |
+
"learning_rate": 0.00010194596887510275,
|
| 582 |
+
"loss": 0.1698,
|
| 583 |
+
"step": 37000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.4912908613696017,
|
| 587 |
+
"grad_norm": 1.2367908954620361,
|
| 588 |
+
"learning_rate": 0.00010062037699832975,
|
| 589 |
+
"loss": 0.174,
|
| 590 |
+
"step": 37500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.5111747395211963,
|
| 594 |
+
"grad_norm": 1.8972302675247192,
|
| 595 |
+
"learning_rate": 9.929743630531032e-05,
|
| 596 |
+
"loss": 0.174,
|
| 597 |
+
"step": 38000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.531058617672791,
|
| 601 |
+
"grad_norm": 1.862426996231079,
|
| 602 |
+
"learning_rate": 9.797184442853735e-05,
|
| 603 |
+
"loss": 0.1713,
|
| 604 |
+
"step": 38500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.5509424958243856,
|
| 608 |
+
"grad_norm": 0.6830955743789673,
|
| 609 |
+
"learning_rate": 9.664625255176437e-05,
|
| 610 |
+
"loss": 0.1676,
|
| 611 |
+
"step": 39000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.5708263739759802,
|
| 615 |
+
"grad_norm": 1.9133577346801758,
|
| 616 |
+
"learning_rate": 9.532066067499139e-05,
|
| 617 |
+
"loss": 0.1715,
|
| 618 |
+
"step": 39500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.5907102521275749,
|
| 622 |
+
"grad_norm": 2.2578415870666504,
|
| 623 |
+
"learning_rate": 9.399506879821841e-05,
|
| 624 |
+
"loss": 0.1692,
|
| 625 |
+
"step": 40000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.6002545136403405,
|
| 629 |
+
"eval_loss": 0.17579330503940582,
|
| 630 |
+
"eval_runtime": 34.0311,
|
| 631 |
+
"eval_samples_per_second": 14.692,
|
| 632 |
+
"eval_steps_per_second": 7.346,
|
| 633 |
+
"step": 40240
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.6105941302791695,
|
| 637 |
+
"grad_norm": 2.9348762035369873,
|
| 638 |
+
"learning_rate": 9.266947692144543e-05,
|
| 639 |
+
"loss": 0.1623,
|
| 640 |
+
"step": 40500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.6304780084307642,
|
| 644 |
+
"grad_norm": 1.4886940717697144,
|
| 645 |
+
"learning_rate": 9.134653622842599e-05,
|
| 646 |
+
"loss": 0.1697,
|
| 647 |
+
"step": 41000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.650361886582359,
|
| 651 |
+
"grad_norm": 4.393771648406982,
|
| 652 |
+
"learning_rate": 9.002094435165301e-05,
|
| 653 |
+
"loss": 0.1704,
|
| 654 |
+
"step": 41500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.6702457647339537,
|
| 658 |
+
"grad_norm": 2.1238715648651123,
|
| 659 |
+
"learning_rate": 8.869535247488003e-05,
|
| 660 |
+
"loss": 0.168,
|
| 661 |
+
"step": 42000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.6901296428855483,
|
| 665 |
+
"grad_norm": 1.604225754737854,
|
| 666 |
+
"learning_rate": 8.736976059810706e-05,
|
| 667 |
+
"loss": 0.1695,
|
| 668 |
+
"step": 42500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.7100135210371432,
|
| 672 |
+
"grad_norm": 2.205331802368164,
|
| 673 |
+
"learning_rate": 8.604416872133408e-05,
|
| 674 |
+
"loss": 0.1713,
|
| 675 |
+
"step": 43000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.7298973991887379,
|
| 679 |
+
"grad_norm": 2.6028106212615967,
|
| 680 |
+
"learning_rate": 8.472122802831465e-05,
|
| 681 |
+
"loss": 0.1696,
|
| 682 |
+
"step": 43500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.7497812773403325,
|
| 686 |
+
"grad_norm": 1.416176199913025,
|
| 687 |
+
"learning_rate": 8.339563615154167e-05,
|
| 688 |
+
"loss": 0.1699,
|
| 689 |
+
"step": 44000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.7696651554919272,
|
| 693 |
+
"grad_norm": 1.434804916381836,
|
| 694 |
+
"learning_rate": 8.207004427476868e-05,
|
| 695 |
+
"loss": 0.1651,
|
| 696 |
+
"step": 44500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.7895490336435218,
|
| 700 |
+
"grad_norm": 1.455796241760254,
|
| 701 |
+
"learning_rate": 8.07444523979957e-05,
|
| 702 |
+
"loss": 0.1559,
|
| 703 |
+
"step": 45000
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.8002863278453831,
|
| 707 |
+
"eval_loss": 0.16793328523635864,
|
| 708 |
+
"eval_runtime": 33.6241,
|
| 709 |
+
"eval_samples_per_second": 14.87,
|
| 710 |
+
"eval_steps_per_second": 7.435,
|
| 711 |
+
"step": 45270
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.8094329117951164,
|
| 715 |
+
"grad_norm": 1.8090940713882446,
|
| 716 |
+
"learning_rate": 7.941886052122272e-05,
|
| 717 |
+
"loss": 0.159,
|
| 718 |
+
"step": 45500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.829316789946711,
|
| 722 |
+
"grad_norm": 0.6885485053062439,
|
| 723 |
+
"learning_rate": 7.80959198282033e-05,
|
| 724 |
+
"loss": 0.1558,
|
| 725 |
+
"step": 46000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.8492006680983057,
|
| 729 |
+
"grad_norm": 1.964609146118164,
|
| 730 |
+
"learning_rate": 7.677032795143032e-05,
|
| 731 |
+
"loss": 0.1681,
|
| 732 |
+
"step": 46500
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.8690845462499006,
|
| 736 |
+
"grad_norm": 1.272363305091858,
|
| 737 |
+
"learning_rate": 7.544473607465734e-05,
|
| 738 |
+
"loss": 0.1634,
|
| 739 |
+
"step": 47000
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.8889684244014953,
|
| 743 |
+
"grad_norm": 2.018178939819336,
|
| 744 |
+
"learning_rate": 7.411914419788436e-05,
|
| 745 |
+
"loss": 0.1613,
|
| 746 |
+
"step": 47500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.90885230255309,
|
| 750 |
+
"grad_norm": 1.9552594423294067,
|
| 751 |
+
"learning_rate": 7.279620350486492e-05,
|
| 752 |
+
"loss": 0.1542,
|
| 753 |
+
"step": 48000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.9287361807046848,
|
| 757 |
+
"grad_norm": 0.6383972764015198,
|
| 758 |
+
"learning_rate": 7.147061162809194e-05,
|
| 759 |
+
"loss": 0.1652,
|
| 760 |
+
"step": 48500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.9486200588562794,
|
| 764 |
+
"grad_norm": 1.6994730234146118,
|
| 765 |
+
"learning_rate": 7.014501975131896e-05,
|
| 766 |
+
"loss": 0.1595,
|
| 767 |
+
"step": 49000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.968503937007874,
|
| 771 |
+
"grad_norm": 2.7284016609191895,
|
| 772 |
+
"learning_rate": 6.881942787454598e-05,
|
| 773 |
+
"loss": 0.1577,
|
| 774 |
+
"step": 49500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.9883878151594687,
|
| 778 |
+
"grad_norm": 1.450149416923523,
|
| 779 |
+
"learning_rate": 6.749383599777301e-05,
|
| 780 |
+
"loss": 0.1624,
|
| 781 |
+
"step": 50000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.0003181420504257,
|
| 785 |
+
"eval_loss": 0.16452698409557343,
|
| 786 |
+
"eval_runtime": 33.506,
|
| 787 |
+
"eval_samples_per_second": 14.923,
|
| 788 |
+
"eval_steps_per_second": 7.461,
|
| 789 |
+
"step": 50300
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 2.0082716933110634,
|
| 793 |
+
"grad_norm": 2.1276376247406006,
|
| 794 |
+
"learning_rate": 6.617089530475358e-05,
|
| 795 |
+
"loss": 0.1458,
|
| 796 |
+
"step": 50500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 2.028155571462658,
|
| 800 |
+
"grad_norm": 1.5807527303695679,
|
| 801 |
+
"learning_rate": 6.48453034279806e-05,
|
| 802 |
+
"loss": 0.1367,
|
| 803 |
+
"step": 51000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 2.0480394496142527,
|
| 807 |
+
"grad_norm": 1.8037587404251099,
|
| 808 |
+
"learning_rate": 6.351971155120762e-05,
|
| 809 |
+
"loss": 0.1394,
|
| 810 |
+
"step": 51500
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 2.0679233277658473,
|
| 814 |
+
"grad_norm": 0.7826844453811646,
|
| 815 |
+
"learning_rate": 6.219411967443464e-05,
|
| 816 |
+
"loss": 0.1357,
|
| 817 |
+
"step": 52000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 2.087807205917442,
|
| 821 |
+
"grad_norm": 2.5608022212982178,
|
| 822 |
+
"learning_rate": 6.0868527797661656e-05,
|
| 823 |
+
"loss": 0.1345,
|
| 824 |
+
"step": 52500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.107691084069037,
|
| 828 |
+
"grad_norm": 1.045422077178955,
|
| 829 |
+
"learning_rate": 5.954558710464223e-05,
|
| 830 |
+
"loss": 0.136,
|
| 831 |
+
"step": 53000
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 2.1275749622206317,
|
| 835 |
+
"grad_norm": 2.0140347480773926,
|
| 836 |
+
"learning_rate": 5.821999522786925e-05,
|
| 837 |
+
"loss": 0.1386,
|
| 838 |
+
"step": 53500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 2.1474588403722263,
|
| 842 |
+
"grad_norm": 2.6173617839813232,
|
| 843 |
+
"learning_rate": 5.689440335109627e-05,
|
| 844 |
+
"loss": 0.1401,
|
| 845 |
+
"step": 54000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 2.167342718523821,
|
| 849 |
+
"grad_norm": 1.4525425434112549,
|
| 850 |
+
"learning_rate": 5.556881147432329e-05,
|
| 851 |
+
"loss": 0.1355,
|
| 852 |
+
"step": 54500
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 2.1872265966754156,
|
| 856 |
+
"grad_norm": 1.2328760623931885,
|
| 857 |
+
"learning_rate": 5.424321959755031e-05,
|
| 858 |
+
"loss": 0.1233,
|
| 859 |
+
"step": 55000
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 2.2003499562554683,
|
| 863 |
+
"eval_loss": 0.15997442603111267,
|
| 864 |
+
"eval_runtime": 32.8121,
|
| 865 |
+
"eval_samples_per_second": 15.238,
|
| 866 |
+
"eval_steps_per_second": 7.619,
|
| 867 |
+
"step": 55330
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.2071104748270103,
|
| 871 |
+
"grad_norm": 1.4365465641021729,
|
| 872 |
+
"learning_rate": 5.291762772077733e-05,
|
| 873 |
+
"loss": 0.1366,
|
| 874 |
+
"step": 55500
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.226994352978605,
|
| 878 |
+
"grad_norm": 2.6288044452667236,
|
| 879 |
+
"learning_rate": 5.1594687027757893e-05,
|
| 880 |
+
"loss": 0.1324,
|
| 881 |
+
"step": 56000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.2468782311301996,
|
| 885 |
+
"grad_norm": 1.7624181509017944,
|
| 886 |
+
"learning_rate": 5.026909515098491e-05,
|
| 887 |
+
"loss": 0.138,
|
| 888 |
+
"step": 56500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 2.2667621092817942,
|
| 892 |
+
"grad_norm": 4.451639175415039,
|
| 893 |
+
"learning_rate": 4.894350327421194e-05,
|
| 894 |
+
"loss": 0.1381,
|
| 895 |
+
"step": 57000
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 2.286645987433389,
|
| 899 |
+
"grad_norm": 1.2079836130142212,
|
| 900 |
+
"learning_rate": 4.761791139743896e-05,
|
| 901 |
+
"loss": 0.1403,
|
| 902 |
+
"step": 57500
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 2.3065298655849835,
|
| 906 |
+
"grad_norm": 1.6635743379592896,
|
| 907 |
+
"learning_rate": 4.629231952066598e-05,
|
| 908 |
+
"loss": 0.1354,
|
| 909 |
+
"step": 58000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 2.3264137437365786,
|
| 913 |
+
"grad_norm": 2.777181625366211,
|
| 914 |
+
"learning_rate": 4.4966727643893005e-05,
|
| 915 |
+
"loss": 0.1349,
|
| 916 |
+
"step": 58500
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 2.3462976218881733,
|
| 920 |
+
"grad_norm": 5.82150936126709,
|
| 921 |
+
"learning_rate": 4.3641135767120024e-05,
|
| 922 |
+
"loss": 0.1416,
|
| 923 |
+
"step": 59000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 2.366181500039768,
|
| 927 |
+
"grad_norm": 1.459591031074524,
|
| 928 |
+
"learning_rate": 4.2315543890347044e-05,
|
| 929 |
+
"loss": 0.1332,
|
| 930 |
+
"step": 59500
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 2.3860653781913626,
|
| 934 |
+
"grad_norm": 1.0342719554901123,
|
| 935 |
+
"learning_rate": 4.0992603197327614e-05,
|
| 936 |
+
"loss": 0.1206,
|
| 937 |
+
"step": 60000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 2.400381770460511,
|
| 941 |
+
"eval_loss": 0.15386110544204712,
|
| 942 |
+
"eval_runtime": 32.525,
|
| 943 |
+
"eval_samples_per_second": 15.373,
|
| 944 |
+
"eval_steps_per_second": 7.686,
|
| 945 |
+
"step": 60360
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 2.405949256342957,
|
| 949 |
+
"grad_norm": 3.423184394836426,
|
| 950 |
+
"learning_rate": 3.966701132055463e-05,
|
| 951 |
+
"loss": 0.1316,
|
| 952 |
+
"step": 60500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 2.425833134494552,
|
| 956 |
+
"grad_norm": 6.525878429412842,
|
| 957 |
+
"learning_rate": 3.834141944378165e-05,
|
| 958 |
+
"loss": 0.1236,
|
| 959 |
+
"step": 61000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 2.4457170126461465,
|
| 963 |
+
"grad_norm": 3.186624765396118,
|
| 964 |
+
"learning_rate": 3.701582756700867e-05,
|
| 965 |
+
"loss": 0.1256,
|
| 966 |
+
"step": 61500
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 2.465600890797741,
|
| 970 |
+
"grad_norm": 1.440810203552246,
|
| 971 |
+
"learning_rate": 3.569023569023569e-05,
|
| 972 |
+
"loss": 0.1371,
|
| 973 |
+
"step": 62000
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 2.485484768949336,
|
| 977 |
+
"grad_norm": 1.9754679203033447,
|
| 978 |
+
"learning_rate": 3.436729499721626e-05,
|
| 979 |
+
"loss": 0.1298,
|
| 980 |
+
"step": 62500
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 2.5053686471009304,
|
| 984 |
+
"grad_norm": 1.1625438928604126,
|
| 985 |
+
"learning_rate": 3.304170312044328e-05,
|
| 986 |
+
"loss": 0.1288,
|
| 987 |
+
"step": 63000
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 2.525252525252525,
|
| 991 |
+
"grad_norm": 1.642461895942688,
|
| 992 |
+
"learning_rate": 3.17161112436703e-05,
|
| 993 |
+
"loss": 0.1299,
|
| 994 |
+
"step": 63500
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 2.54513640340412,
|
| 998 |
+
"grad_norm": 2.076730728149414,
|
| 999 |
+
"learning_rate": 3.0390519366897317e-05,
|
| 1000 |
+
"loss": 0.1282,
|
| 1001 |
+
"step": 64000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 2.5650202815557144,
|
| 1005 |
+
"grad_norm": 1.2230627536773682,
|
| 1006 |
+
"learning_rate": 2.9064927490124343e-05,
|
| 1007 |
+
"loss": 0.123,
|
| 1008 |
+
"step": 64500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 2.5849041597073095,
|
| 1012 |
+
"grad_norm": 2.4651546478271484,
|
| 1013 |
+
"learning_rate": 2.7741986797104907e-05,
|
| 1014 |
+
"loss": 0.1296,
|
| 1015 |
+
"step": 65000
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 2.600413584665553,
|
| 1019 |
+
"eval_loss": 0.15161803364753723,
|
| 1020 |
+
"eval_runtime": 32.4931,
|
| 1021 |
+
"eval_samples_per_second": 15.388,
|
| 1022 |
+
"eval_steps_per_second": 7.694,
|
| 1023 |
+
"step": 65390
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 2.604788037858904,
|
| 1027 |
+
"grad_norm": 1.16021728515625,
|
| 1028 |
+
"learning_rate": 2.6416394920331926e-05,
|
| 1029 |
+
"loss": 0.1306,
|
| 1030 |
+
"step": 65500
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 2.6246719160104988,
|
| 1034 |
+
"grad_norm": 1.5166651010513306,
|
| 1035 |
+
"learning_rate": 2.5090803043558952e-05,
|
| 1036 |
+
"loss": 0.1292,
|
| 1037 |
+
"step": 66000
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.6445557941620934,
|
| 1041 |
+
"grad_norm": 2.6024417877197266,
|
| 1042 |
+
"learning_rate": 2.3765211166785972e-05,
|
| 1043 |
+
"loss": 0.1248,
|
| 1044 |
+
"step": 66500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.664439672313688,
|
| 1048 |
+
"grad_norm": 1.9998677968978882,
|
| 1049 |
+
"learning_rate": 2.243961929001299e-05,
|
| 1050 |
+
"loss": 0.1255,
|
| 1051 |
+
"step": 67000
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 2.6843235504652827,
|
| 1055 |
+
"grad_norm": 1.2825827598571777,
|
| 1056 |
+
"learning_rate": 2.1116678596993558e-05,
|
| 1057 |
+
"loss": 0.1346,
|
| 1058 |
+
"step": 67500
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 2.7042074286168774,
|
| 1062 |
+
"grad_norm": 1.9823710918426514,
|
| 1063 |
+
"learning_rate": 1.9793737903974125e-05,
|
| 1064 |
+
"loss": 0.124,
|
| 1065 |
+
"step": 68000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 2.724091306768472,
|
| 1069 |
+
"grad_norm": 0.2522190511226654,
|
| 1070 |
+
"learning_rate": 1.8468146027201148e-05,
|
| 1071 |
+
"loss": 0.1282,
|
| 1072 |
+
"step": 68500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 2.7439751849200666,
|
| 1076 |
+
"grad_norm": 2.7154924869537354,
|
| 1077 |
+
"learning_rate": 1.7142554150428167e-05,
|
| 1078 |
+
"loss": 0.1196,
|
| 1079 |
+
"step": 69000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 2.7638590630716617,
|
| 1083 |
+
"grad_norm": 0.9731872081756592,
|
| 1084 |
+
"learning_rate": 1.5816962273655187e-05,
|
| 1085 |
+
"loss": 0.125,
|
| 1086 |
+
"step": 69500
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 2.783742941223256,
|
| 1090 |
+
"grad_norm": 2.066401481628418,
|
| 1091 |
+
"learning_rate": 1.4491370396882208e-05,
|
| 1092 |
+
"loss": 0.1277,
|
| 1093 |
+
"step": 70000
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 2.800445398870596,
|
| 1097 |
+
"eval_loss": 0.14814555644989014,
|
| 1098 |
+
"eval_runtime": 33.4698,
|
| 1099 |
+
"eval_samples_per_second": 14.939,
|
| 1100 |
+
"eval_steps_per_second": 7.469,
|
| 1101 |
+
"step": 70420
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.803626819374851,
|
| 1105 |
+
"grad_norm": 0.9782494306564331,
|
| 1106 |
+
"learning_rate": 1.316577852010923e-05,
|
| 1107 |
+
"loss": 0.1197,
|
| 1108 |
+
"step": 70500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.8235106975264457,
|
| 1112 |
+
"grad_norm": 2.0531044006347656,
|
| 1113 |
+
"learning_rate": 1.184018664333625e-05,
|
| 1114 |
+
"loss": 0.122,
|
| 1115 |
+
"step": 71000
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.8433945756780403,
|
| 1119 |
+
"grad_norm": 3.9490396976470947,
|
| 1120 |
+
"learning_rate": 1.0514594766563272e-05,
|
| 1121 |
+
"loss": 0.1198,
|
| 1122 |
+
"step": 71500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.863278453829635,
|
| 1126 |
+
"grad_norm": 1.4943678379058838,
|
| 1127 |
+
"learning_rate": 9.189002889790293e-06,
|
| 1128 |
+
"loss": 0.1252,
|
| 1129 |
+
"step": 72000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.8831623319812296,
|
| 1133 |
+
"grad_norm": 0.026091307401657104,
|
| 1134 |
+
"learning_rate": 7.863411013017313e-06,
|
| 1135 |
+
"loss": 0.123,
|
| 1136 |
+
"step": 72500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.9030462101328243,
|
| 1140 |
+
"grad_norm": 0.7524722814559937,
|
| 1141 |
+
"learning_rate": 6.540470319997879e-06,
|
| 1142 |
+
"loss": 0.1202,
|
| 1143 |
+
"step": 73000
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.922930088284419,
|
| 1147 |
+
"grad_norm": 4.17079496383667,
|
| 1148 |
+
"learning_rate": 5.2148784432249e-06,
|
| 1149 |
+
"loss": 0.1163,
|
| 1150 |
+
"step": 73500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.9428139664360136,
|
| 1154 |
+
"grad_norm": 1.407780647277832,
|
| 1155 |
+
"learning_rate": 3.889286566451921e-06,
|
| 1156 |
+
"loss": 0.1209,
|
| 1157 |
+
"step": 74000
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.962697844587608,
|
| 1161 |
+
"grad_norm": 3.6621692180633545,
|
| 1162 |
+
"learning_rate": 2.563694689678942e-06,
|
| 1163 |
+
"loss": 0.1283,
|
| 1164 |
+
"step": 74500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.9825817227392033,
|
| 1168 |
+
"grad_norm": 2.399603843688965,
|
| 1169 |
+
"learning_rate": 1.2381028129059626e-06,
|
| 1170 |
+
"loss": 0.1163,
|
| 1171 |
+
"step": 75000
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.0,
|
| 1175 |
+
"step": 75438,
|
| 1176 |
+
"total_flos": 4.793250966804726e+17,
|
| 1177 |
+
"train_loss": 0.18797989167699114,
|
| 1178 |
+
"train_runtime": 40772.1174,
|
| 1179 |
+
"train_samples_per_second": 3.7,
|
| 1180 |
+
"train_steps_per_second": 1.85
|
| 1181 |
}
|
| 1182 |
],
|
| 1183 |
"logging_steps": 500,
|
| 1184 |
+
"max_steps": 75438,
|
| 1185 |
"num_input_tokens_seen": 0,
|
| 1186 |
+
"num_train_epochs": 3,
|
| 1187 |
"save_steps": 500,
|
| 1188 |
"stateful_callbacks": {
|
| 1189 |
"TrainerControl": {
|
|
|
|
| 1197 |
"attributes": {}
|
| 1198 |
}
|
| 1199 |
},
|
| 1200 |
+
"total_flos": 4.793250966804726e+17,
|
| 1201 |
"train_batch_size": 2,
|
| 1202 |
"trial_name": null,
|
| 1203 |
"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 5496
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bc246d7e65a14f84b25357cc28bc2874d826c12da43456453359a4091f554ee2
|
| 3 |
size 5496
|