Model save
Browse files- README.md +22 -22
- all_results.json +16 -0
- eval_results.json +10 -0
- model.safetensors +1 -1
- special_tokens_map.json +28 -4
- tokenizer.json +0 -0
- train_results.json +9 -0
- trainer_state.json +476 -0
- training_args.bin +1 -1
- vocab.json +0 -0
README.md
CHANGED
|
@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model was trained from scratch on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 3.
|
| 20 |
-
- Accuracy: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
@@ -52,26 +52,26 @@ The following hyperparameters were used during training:
|
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 54 |
|:-------------:|:-------:|:-----:|:---------------:|:--------:|
|
| 55 |
-
| 6.
|
| 56 |
-
| 4.
|
| 57 |
-
| 3.
|
| 58 |
-
| 3.
|
| 59 |
-
| 3.
|
| 60 |
-
| 3.
|
| 61 |
-
| 3.
|
| 62 |
-
| 3.
|
| 63 |
-
| 3.
|
| 64 |
-
| 3.
|
| 65 |
-
| 2.
|
| 66 |
-
| 2.
|
| 67 |
-
| 2.
|
| 68 |
-
| 2.
|
| 69 |
-
| 2.
|
| 70 |
-
| 2.
|
| 71 |
-
| 2.
|
| 72 |
-
| 2.
|
| 73 |
-
| 2.
|
| 74 |
-
| 2.
|
| 75 |
|
| 76 |
|
| 77 |
### Framework versions
|
|
|
|
| 16 |
|
| 17 |
This model was trained from scratch on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 3.0136
|
| 20 |
+
- Accuracy: 0.4207
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 54 |
|:-------------:|:-------:|:-----:|:---------------:|:--------:|
|
| 55 |
+
| 6.1723 | 0.9999 | 1788 | 4.2504 | 0.3067 |
|
| 56 |
+
| 4.0454 | 1.9999 | 3576 | 3.7300 | 0.3471 |
|
| 57 |
+
| 3.6164 | 2.9998 | 5364 | 3.4747 | 0.3703 |
|
| 58 |
+
| 3.394 | 3.9997 | 7152 | 3.3376 | 0.3834 |
|
| 59 |
+
| 3.3055 | 4.9997 | 8940 | 3.2589 | 0.3908 |
|
| 60 |
+
| 3.2018 | 5.9996 | 10728 | 3.2086 | 0.3957 |
|
| 61 |
+
| 3.1366 | 6.9995 | 12516 | 3.1760 | 0.3991 |
|
| 62 |
+
| 3.093 | 8.0 | 14305 | 3.1520 | 0.4016 |
|
| 63 |
+
| 3.0613 | 8.9999 | 16093 | 3.1356 | 0.4037 |
|
| 64 |
+
| 3.0161 | 9.9999 | 17881 | 3.1245 | 0.4046 |
|
| 65 |
+
| 2.9952 | 10.9998 | 19669 | 3.1162 | 0.4061 |
|
| 66 |
+
| 2.9865 | 11.9997 | 21457 | 3.1086 | 0.4065 |
|
| 67 |
+
| 2.9786 | 12.9997 | 23245 | 3.1062 | 0.4073 |
|
| 68 |
+
| 2.9743 | 13.9996 | 25033 | 3.1010 | 0.4076 |
|
| 69 |
+
| 2.9328 | 14.9995 | 26821 | 3.0967 | 0.4084 |
|
| 70 |
+
| 2.9369 | 16.0 | 28610 | 3.0958 | 0.4086 |
|
| 71 |
+
| 2.9408 | 16.9999 | 30398 | 3.0929 | 0.4088 |
|
| 72 |
+
| 2.9445 | 17.9999 | 32186 | 3.0917 | 0.4094 |
|
| 73 |
+
| 2.8906 | 18.9998 | 33974 | 3.0351 | 0.4163 |
|
| 74 |
+
| 2.7359 | 19.9986 | 35760 | 3.0136 | 0.4207 |
|
| 75 |
|
| 76 |
|
| 77 |
### Framework versions
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.99718982717437,
|
| 3 |
+
"eval_accuracy": 0.4210676920791576,
|
| 4 |
+
"eval_loss": 3.009352207183838,
|
| 5 |
+
"eval_runtime": 128.3569,
|
| 6 |
+
"eval_samples": 45143,
|
| 7 |
+
"eval_samples_per_second": 351.699,
|
| 8 |
+
"eval_steps_per_second": 5.5,
|
| 9 |
+
"perplexity": 20.274262149048695,
|
| 10 |
+
"total_flos": 1.18991215558656e+18,
|
| 11 |
+
"train_loss": 3.2179960425614373,
|
| 12 |
+
"train_runtime": 62492.0089,
|
| 13 |
+
"train_samples": 455458,
|
| 14 |
+
"train_samples_per_second": 145.765,
|
| 15 |
+
"train_steps_per_second": 0.569
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.99718982717437,
|
| 3 |
+
"eval_accuracy": 0.4210676920791576,
|
| 4 |
+
"eval_loss": 3.009352207183838,
|
| 5 |
+
"eval_runtime": 128.3569,
|
| 6 |
+
"eval_samples": 45143,
|
| 7 |
+
"eval_samples_per_second": 351.699,
|
| 8 |
+
"eval_steps_per_second": 5.5,
|
| 9 |
+
"perplexity": 20.274262149048695
|
| 10 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 441702288
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:18c4a883223cd42c0cb8a7322fdd660893099fa47fd031f308605de96818f353
|
| 3 |
size 441702288
|
special_tokens_map.json
CHANGED
|
@@ -1,6 +1,30 @@
|
|
| 1 |
{
|
| 2 |
-
"bos_token":
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 6 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<pad>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
}
|
tokenizer.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.99718982717437,
|
| 3 |
+
"total_flos": 1.18991215558656e+18,
|
| 4 |
+
"train_loss": 3.2179960425614373,
|
| 5 |
+
"train_runtime": 62492.0089,
|
| 6 |
+
"train_samples": 455458,
|
| 7 |
+
"train_samples_per_second": 145.765,
|
| 8 |
+
"train_steps_per_second": 0.569
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,476 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 3.009352207183838,
|
| 3 |
+
"best_model_checkpoint": "models/babylm-unablated_seed-42_1e-3/checkpoint-35580",
|
| 4 |
+
"epoch": 19.99718982717437,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 35580,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.5620345651257552,
|
| 13 |
+
"grad_norm": 0.747050404548645,
|
| 14 |
+
"learning_rate": 3.125e-05,
|
| 15 |
+
"loss": 6.1639,
|
| 16 |
+
"step": 1000
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.9998594913587185,
|
| 20 |
+
"eval_accuracy": 0.3043492726599091,
|
| 21 |
+
"eval_loss": 4.264366149902344,
|
| 22 |
+
"eval_runtime": 127.2817,
|
| 23 |
+
"eval_samples_per_second": 354.67,
|
| 24 |
+
"eval_steps_per_second": 5.547,
|
| 25 |
+
"step": 1779
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.1240691302515105,
|
| 29 |
+
"grad_norm": 0.6358739733695984,
|
| 30 |
+
"learning_rate": 6.25e-05,
|
| 31 |
+
"loss": 4.4584,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.6861036953772657,
|
| 36 |
+
"grad_norm": 0.710475742816925,
|
| 37 |
+
"learning_rate": 9.375e-05,
|
| 38 |
+
"loss": 4.0418,
|
| 39 |
+
"step": 3000
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.999718982717437,
|
| 43 |
+
"eval_accuracy": 0.34673388660782967,
|
| 44 |
+
"eval_loss": 3.7366583347320557,
|
| 45 |
+
"eval_runtime": 128.0547,
|
| 46 |
+
"eval_samples_per_second": 352.529,
|
| 47 |
+
"eval_steps_per_second": 5.513,
|
| 48 |
+
"step": 3558
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 2.248138260503021,
|
| 52 |
+
"grad_norm": 0.6259163022041321,
|
| 53 |
+
"learning_rate": 0.000125,
|
| 54 |
+
"loss": 3.7842,
|
| 55 |
+
"step": 4000
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 2.810172825628776,
|
| 59 |
+
"grad_norm": 0.5914742350578308,
|
| 60 |
+
"learning_rate": 0.00015625,
|
| 61 |
+
"loss": 3.6074,
|
| 62 |
+
"step": 5000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.9995784740761557,
|
| 66 |
+
"eval_accuracy": 0.3700874736621273,
|
| 67 |
+
"eval_loss": 3.4760947227478027,
|
| 68 |
+
"eval_runtime": 128.4007,
|
| 69 |
+
"eval_samples_per_second": 351.579,
|
| 70 |
+
"eval_steps_per_second": 5.498,
|
| 71 |
+
"step": 5337
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 3.3722073907545314,
|
| 75 |
+
"grad_norm": 0.532345712184906,
|
| 76 |
+
"learning_rate": 0.0001875,
|
| 77 |
+
"loss": 3.4689,
|
| 78 |
+
"step": 6000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 3.9342419558802866,
|
| 82 |
+
"grad_norm": 0.49571847915649414,
|
| 83 |
+
"learning_rate": 0.00021875,
|
| 84 |
+
"loss": 3.3836,
|
| 85 |
+
"step": 7000
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 4.0,
|
| 89 |
+
"eval_accuracy": 0.38321377861114986,
|
| 90 |
+
"eval_loss": 3.33798885345459,
|
| 91 |
+
"eval_runtime": 127.867,
|
| 92 |
+
"eval_samples_per_second": 353.047,
|
| 93 |
+
"eval_steps_per_second": 5.521,
|
| 94 |
+
"step": 7117
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 4.496276521006042,
|
| 98 |
+
"grad_norm": 0.4585084319114685,
|
| 99 |
+
"learning_rate": 0.00025,
|
| 100 |
+
"loss": 3.2939,
|
| 101 |
+
"step": 8000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 4.999859491358719,
|
| 105 |
+
"eval_accuracy": 0.3912746987459893,
|
| 106 |
+
"eval_loss": 3.2569355964660645,
|
| 107 |
+
"eval_runtime": 128.3798,
|
| 108 |
+
"eval_samples_per_second": 351.636,
|
| 109 |
+
"eval_steps_per_second": 5.499,
|
| 110 |
+
"step": 8896
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 5.0583110861317975,
|
| 114 |
+
"grad_norm": 0.3779755234718323,
|
| 115 |
+
"learning_rate": 0.00028125000000000003,
|
| 116 |
+
"loss": 3.2499,
|
| 117 |
+
"step": 9000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 5.620345651257552,
|
| 121 |
+
"grad_norm": 0.3809504210948944,
|
| 122 |
+
"learning_rate": 0.0003125,
|
| 123 |
+
"loss": 3.1911,
|
| 124 |
+
"step": 10000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 5.999718982717437,
|
| 128 |
+
"eval_accuracy": 0.3958713334923053,
|
| 129 |
+
"eval_loss": 3.208075523376465,
|
| 130 |
+
"eval_runtime": 128.1646,
|
| 131 |
+
"eval_samples_per_second": 352.227,
|
| 132 |
+
"eval_steps_per_second": 5.509,
|
| 133 |
+
"step": 10675
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 6.182380216383308,
|
| 137 |
+
"grad_norm": 0.361873596906662,
|
| 138 |
+
"learning_rate": 0.00034375,
|
| 139 |
+
"loss": 3.1594,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 6.744414781509063,
|
| 144 |
+
"grad_norm": 0.3257278501987457,
|
| 145 |
+
"learning_rate": 0.000375,
|
| 146 |
+
"loss": 3.1276,
|
| 147 |
+
"step": 12000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 6.999578474076156,
|
| 151 |
+
"eval_accuracy": 0.3993205903853245,
|
| 152 |
+
"eval_loss": 3.174260377883911,
|
| 153 |
+
"eval_runtime": 128.5693,
|
| 154 |
+
"eval_samples_per_second": 351.118,
|
| 155 |
+
"eval_steps_per_second": 5.491,
|
| 156 |
+
"step": 12454
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 7.306449346634818,
|
| 160 |
+
"grad_norm": 0.3252241015434265,
|
| 161 |
+
"learning_rate": 0.0004061875,
|
| 162 |
+
"loss": 3.0935,
|
| 163 |
+
"step": 13000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 7.868483911760573,
|
| 167 |
+
"grad_norm": 0.31484290957450867,
|
| 168 |
+
"learning_rate": 0.0004374375,
|
| 169 |
+
"loss": 3.0834,
|
| 170 |
+
"step": 14000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 8.0,
|
| 174 |
+
"eval_accuracy": 0.4013973026022318,
|
| 175 |
+
"eval_loss": 3.1533353328704834,
|
| 176 |
+
"eval_runtime": 128.32,
|
| 177 |
+
"eval_samples_per_second": 351.8,
|
| 178 |
+
"eval_steps_per_second": 5.502,
|
| 179 |
+
"step": 14234
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 8.430518476886329,
|
| 183 |
+
"grad_norm": 0.31745150685310364,
|
| 184 |
+
"learning_rate": 0.0004686875,
|
| 185 |
+
"loss": 3.0435,
|
| 186 |
+
"step": 15000
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 8.992553042012084,
|
| 190 |
+
"grad_norm": 0.2775208353996277,
|
| 191 |
+
"learning_rate": 0.0004999375,
|
| 192 |
+
"loss": 3.0525,
|
| 193 |
+
"step": 16000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 8.999859491358718,
|
| 197 |
+
"eval_accuracy": 0.4032936728730878,
|
| 198 |
+
"eval_loss": 3.1344425678253174,
|
| 199 |
+
"eval_runtime": 128.3823,
|
| 200 |
+
"eval_samples_per_second": 351.629,
|
| 201 |
+
"eval_steps_per_second": 5.499,
|
| 202 |
+
"step": 16013
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 9.554587607137838,
|
| 206 |
+
"grad_norm": 0.28058525919914246,
|
| 207 |
+
"learning_rate": 0.00053115625,
|
| 208 |
+
"loss": 3.0036,
|
| 209 |
+
"step": 17000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 9.999718982717438,
|
| 213 |
+
"eval_accuracy": 0.405010656766971,
|
| 214 |
+
"eval_loss": 3.122068405151367,
|
| 215 |
+
"eval_runtime": 128.3587,
|
| 216 |
+
"eval_samples_per_second": 351.694,
|
| 217 |
+
"eval_steps_per_second": 5.5,
|
| 218 |
+
"step": 17792
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 10.116622172263595,
|
| 222 |
+
"grad_norm": 0.2726403772830963,
|
| 223 |
+
"learning_rate": 0.00056240625,
|
| 224 |
+
"loss": 3.0137,
|
| 225 |
+
"step": 18000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 10.67865673738935,
|
| 229 |
+
"grad_norm": 0.255198210477829,
|
| 230 |
+
"learning_rate": 0.000593625,
|
| 231 |
+
"loss": 2.9901,
|
| 232 |
+
"step": 19000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 10.999578474076156,
|
| 236 |
+
"eval_accuracy": 0.4056058894328394,
|
| 237 |
+
"eval_loss": 3.1155996322631836,
|
| 238 |
+
"eval_runtime": 128.3954,
|
| 239 |
+
"eval_samples_per_second": 351.594,
|
| 240 |
+
"eval_steps_per_second": 5.499,
|
| 241 |
+
"step": 19571
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 11.240691302515105,
|
| 245 |
+
"grad_norm": 0.26096227765083313,
|
| 246 |
+
"learning_rate": 0.000624875,
|
| 247 |
+
"loss": 2.9801,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 11.80272586764086,
|
| 252 |
+
"grad_norm": 0.2451123744249344,
|
| 253 |
+
"learning_rate": 0.000656125,
|
| 254 |
+
"loss": 2.9798,
|
| 255 |
+
"step": 21000
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 12.0,
|
| 259 |
+
"eval_accuracy": 0.4068902611439986,
|
| 260 |
+
"eval_loss": 3.1083483695983887,
|
| 261 |
+
"eval_runtime": 128.4387,
|
| 262 |
+
"eval_samples_per_second": 351.475,
|
| 263 |
+
"eval_steps_per_second": 5.497,
|
| 264 |
+
"step": 21351
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 12.364760432766616,
|
| 268 |
+
"grad_norm": 0.251653254032135,
|
| 269 |
+
"learning_rate": 0.0006873749999999999,
|
| 270 |
+
"loss": 2.9544,
|
| 271 |
+
"step": 22000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 12.92679499789237,
|
| 275 |
+
"grad_norm": 0.23384283483028412,
|
| 276 |
+
"learning_rate": 0.00071859375,
|
| 277 |
+
"loss": 2.9729,
|
| 278 |
+
"step": 23000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 12.999859491358718,
|
| 282 |
+
"eval_accuracy": 0.40739992694240046,
|
| 283 |
+
"eval_loss": 3.1014933586120605,
|
| 284 |
+
"eval_runtime": 128.3695,
|
| 285 |
+
"eval_samples_per_second": 351.665,
|
| 286 |
+
"eval_steps_per_second": 5.5,
|
| 287 |
+
"step": 23130
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 13.488829563018125,
|
| 291 |
+
"grad_norm": 0.21852266788482666,
|
| 292 |
+
"learning_rate": 0.0007498437500000001,
|
| 293 |
+
"loss": 2.9356,
|
| 294 |
+
"step": 24000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 13.999718982717438,
|
| 298 |
+
"eval_accuracy": 0.40764333644761985,
|
| 299 |
+
"eval_loss": 3.0996627807617188,
|
| 300 |
+
"eval_runtime": 128.4805,
|
| 301 |
+
"eval_samples_per_second": 351.361,
|
| 302 |
+
"eval_steps_per_second": 5.495,
|
| 303 |
+
"step": 24909
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 14.05086412814388,
|
| 307 |
+
"grad_norm": 0.2264845222234726,
|
| 308 |
+
"learning_rate": 0.0007810625,
|
| 309 |
+
"loss": 2.9577,
|
| 310 |
+
"step": 25000
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 14.612898693269637,
|
| 314 |
+
"grad_norm": 0.21793274581432343,
|
| 315 |
+
"learning_rate": 0.0008123125,
|
| 316 |
+
"loss": 2.9282,
|
| 317 |
+
"step": 26000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 14.999578474076156,
|
| 321 |
+
"eval_accuracy": 0.4083614900449248,
|
| 322 |
+
"eval_loss": 3.0948216915130615,
|
| 323 |
+
"eval_runtime": 128.3484,
|
| 324 |
+
"eval_samples_per_second": 351.722,
|
| 325 |
+
"eval_steps_per_second": 5.501,
|
| 326 |
+
"step": 26688
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 15.174933258395392,
|
| 330 |
+
"grad_norm": 0.21455030143260956,
|
| 331 |
+
"learning_rate": 0.00084353125,
|
| 332 |
+
"loss": 2.9359,
|
| 333 |
+
"step": 27000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 15.736967823521146,
|
| 337 |
+
"grad_norm": 0.20253227651119232,
|
| 338 |
+
"learning_rate": 0.00087478125,
|
| 339 |
+
"loss": 2.9332,
|
| 340 |
+
"step": 28000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 16.0,
|
| 344 |
+
"eval_accuracy": 0.40898035132800215,
|
| 345 |
+
"eval_loss": 3.0927815437316895,
|
| 346 |
+
"eval_runtime": 128.2022,
|
| 347 |
+
"eval_samples_per_second": 352.123,
|
| 348 |
+
"eval_steps_per_second": 5.507,
|
| 349 |
+
"step": 28468
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 16.2990023886469,
|
| 353 |
+
"grad_norm": 0.20522606372833252,
|
| 354 |
+
"learning_rate": 0.0009060312499999999,
|
| 355 |
+
"loss": 2.9202,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 16.861036953772658,
|
| 360 |
+
"grad_norm": 0.19934597611427307,
|
| 361 |
+
"learning_rate": 0.00093728125,
|
| 362 |
+
"loss": 2.9346,
|
| 363 |
+
"step": 30000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 16.99985949135872,
|
| 367 |
+
"eval_accuracy": 0.4089585469790335,
|
| 368 |
+
"eval_loss": 3.089961290359497,
|
| 369 |
+
"eval_runtime": 128.5393,
|
| 370 |
+
"eval_samples_per_second": 351.2,
|
| 371 |
+
"eval_steps_per_second": 5.492,
|
| 372 |
+
"step": 30247
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 17.42307151889841,
|
| 376 |
+
"grad_norm": 0.2042018175125122,
|
| 377 |
+
"learning_rate": 0.00096853125,
|
| 378 |
+
"loss": 2.9069,
|
| 379 |
+
"step": 31000
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 17.985106084024167,
|
| 383 |
+
"grad_norm": 0.1865146905183792,
|
| 384 |
+
"learning_rate": 0.00099978125,
|
| 385 |
+
"loss": 2.9394,
|
| 386 |
+
"step": 32000
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 17.999718982717436,
|
| 390 |
+
"eval_accuracy": 0.40965750232485615,
|
| 391 |
+
"eval_loss": 3.087404489517212,
|
| 392 |
+
"eval_runtime": 128.1261,
|
| 393 |
+
"eval_samples_per_second": 352.332,
|
| 394 |
+
"eval_steps_per_second": 5.51,
|
| 395 |
+
"step": 32026
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 18.547140649149924,
|
| 399 |
+
"grad_norm": 0.1902616024017334,
|
| 400 |
+
"learning_rate": 0.0007229050279329609,
|
| 401 |
+
"loss": 2.8743,
|
| 402 |
+
"step": 33000
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 18.999578474076156,
|
| 406 |
+
"eval_accuracy": 0.4162921053054498,
|
| 407 |
+
"eval_loss": 3.0349581241607666,
|
| 408 |
+
"eval_runtime": 128.4097,
|
| 409 |
+
"eval_samples_per_second": 351.554,
|
| 410 |
+
"eval_steps_per_second": 5.498,
|
| 411 |
+
"step": 33805
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 19.109175214275677,
|
| 415 |
+
"grad_norm": 0.20329073071479797,
|
| 416 |
+
"learning_rate": 0.0004435754189944134,
|
| 417 |
+
"loss": 2.8348,
|
| 418 |
+
"step": 34000
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 19.671209779401433,
|
| 422 |
+
"grad_norm": 0.18743157386779785,
|
| 423 |
+
"learning_rate": 0.0001645251396648045,
|
| 424 |
+
"loss": 2.729,
|
| 425 |
+
"step": 35000
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 19.99718982717437,
|
| 429 |
+
"eval_accuracy": 0.4210676920791576,
|
| 430 |
+
"eval_loss": 3.009352207183838,
|
| 431 |
+
"eval_runtime": 129.4424,
|
| 432 |
+
"eval_samples_per_second": 348.75,
|
| 433 |
+
"eval_steps_per_second": 5.454,
|
| 434 |
+
"step": 35580
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 19.99718982717437,
|
| 438 |
+
"step": 35580,
|
| 439 |
+
"total_flos": 1.18991215558656e+18,
|
| 440 |
+
"train_loss": 3.2179960425614373,
|
| 441 |
+
"train_runtime": 62492.0089,
|
| 442 |
+
"train_samples_per_second": 145.765,
|
| 443 |
+
"train_steps_per_second": 0.569
|
| 444 |
+
}
|
| 445 |
+
],
|
| 446 |
+
"logging_steps": 1000,
|
| 447 |
+
"max_steps": 35580,
|
| 448 |
+
"num_input_tokens_seen": 0,
|
| 449 |
+
"num_train_epochs": 20,
|
| 450 |
+
"save_steps": 500,
|
| 451 |
+
"stateful_callbacks": {
|
| 452 |
+
"EarlyStoppingCallback": {
|
| 453 |
+
"args": {
|
| 454 |
+
"early_stopping_patience": 3,
|
| 455 |
+
"early_stopping_threshold": 0.0
|
| 456 |
+
},
|
| 457 |
+
"attributes": {
|
| 458 |
+
"early_stopping_patience_counter": 0
|
| 459 |
+
}
|
| 460 |
+
},
|
| 461 |
+
"TrainerControl": {
|
| 462 |
+
"args": {
|
| 463 |
+
"should_epoch_stop": false,
|
| 464 |
+
"should_evaluate": false,
|
| 465 |
+
"should_log": false,
|
| 466 |
+
"should_save": true,
|
| 467 |
+
"should_training_stop": true
|
| 468 |
+
},
|
| 469 |
+
"attributes": {}
|
| 470 |
+
}
|
| 471 |
+
},
|
| 472 |
+
"total_flos": 1.18991215558656e+18,
|
| 473 |
+
"train_batch_size": 32,
|
| 474 |
+
"trial_name": null,
|
| 475 |
+
"trial_params": null
|
| 476 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5304
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94ea9885287048c764a4765857fc29d5a67c1621d5e8da170add10a216712460
|
| 3 |
size 5304
|
vocab.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|