Commit
•
23c638e
1
Parent(s):
1338531
End of training
Browse files- README.md +42 -14
- config.json +1 -1
- model.safetensors +1 -1
- trainer_state.json +487 -23
- training_args.bin +2 -2
README.md
CHANGED
|
@@ -1,10 +1,11 @@
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
license: apache-2.0
|
| 4 |
-
|
| 5 |
-
- accuracy
|
| 6 |
tags:
|
| 7 |
- generated_from_trainer
|
|
|
|
|
|
|
| 8 |
model-index:
|
| 9 |
- name: wav2vec2-base-960h-EMOPIA
|
| 10 |
results: []
|
|
@@ -17,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss:
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -41,23 +42,50 @@ The following hyperparameters were used during training:
|
|
| 41 |
- train_batch_size: 1
|
| 42 |
- eval_batch_size: 1
|
| 43 |
- seed: 42
|
| 44 |
-
- gradient_accumulation_steps: 3
|
| 45 |
-
- total_train_batch_size: 3
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
-
- num_epochs:
|
| 49 |
- mixed_precision_training: Native AMP
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
-
| Training Loss | Epoch | Step
|
| 54 |
-
|
| 55 |
-
| 1.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
|
| 57 |
|
| 58 |
### Framework versions
|
| 59 |
|
| 60 |
-
- Transformers 4.
|
| 61 |
-
- Pytorch 2.
|
| 62 |
-
- Datasets 2.
|
| 63 |
- Tokenizers 0.19.1
|
|
|
|
| 1 |
---
|
| 2 |
+
library_name: transformers
|
| 3 |
license: apache-2.0
|
| 4 |
+
base_model: facebook/wav2vec2-base-960h
|
|
|
|
| 5 |
tags:
|
| 6 |
- generated_from_trainer
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
model-index:
|
| 10 |
- name: wav2vec2-base-960h-EMOPIA
|
| 11 |
results: []
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 2.4624
|
| 22 |
+
- Accuracy: 0.6620
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 42 |
- train_batch_size: 1
|
| 43 |
- eval_batch_size: 1
|
| 44 |
- seed: 42
|
|
|
|
|
|
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
+
- num_epochs: 30
|
| 48 |
- mixed_precision_training: Native AMP
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 53 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|
|
| 54 |
+
| 1.2041 | 1.0 | 807 | 1.1999 | 0.3662 |
|
| 55 |
+
| 1.067 | 2.0 | 1614 | 1.4613 | 0.4507 |
|
| 56 |
+
| 1.1007 | 3.0 | 2421 | 1.5213 | 0.4085 |
|
| 57 |
+
| 1.1945 | 4.0 | 3228 | 1.6642 | 0.6056 |
|
| 58 |
+
| 1.3665 | 5.0 | 4035 | 2.9908 | 0.4366 |
|
| 59 |
+
| 1.4506 | 6.0 | 4842 | 1.9230 | 0.6056 |
|
| 60 |
+
| 1.495 | 7.0 | 5649 | 1.6813 | 0.6761 |
|
| 61 |
+
| 1.2605 | 8.0 | 6456 | 1.8937 | 0.6620 |
|
| 62 |
+
| 1.2713 | 9.0 | 7263 | 1.6284 | 0.6901 |
|
| 63 |
+
| 1.2608 | 10.0 | 8070 | 1.9438 | 0.6479 |
|
| 64 |
+
| 1.2068 | 11.0 | 8877 | 1.5237 | 0.7183 |
|
| 65 |
+
| 1.0478 | 12.0 | 9684 | 2.0007 | 0.6338 |
|
| 66 |
+
| 1.1282 | 13.0 | 10491 | 1.5307 | 0.7465 |
|
| 67 |
+
| 0.9433 | 14.0 | 11298 | 2.0042 | 0.6479 |
|
| 68 |
+
| 0.9574 | 15.0 | 12105 | 2.1985 | 0.6338 |
|
| 69 |
+
| 0.8737 | 16.0 | 12912 | 2.1568 | 0.6479 |
|
| 70 |
+
| 0.8937 | 17.0 | 13719 | 2.2980 | 0.6197 |
|
| 71 |
+
| 0.8681 | 18.0 | 14526 | 2.3268 | 0.6197 |
|
| 72 |
+
| 0.8005 | 19.0 | 15333 | 2.4827 | 0.6479 |
|
| 73 |
+
| 0.8176 | 20.0 | 16140 | 2.4842 | 0.6338 |
|
| 74 |
+
| 0.8133 | 21.0 | 16947 | 2.0620 | 0.6761 |
|
| 75 |
+
| 0.7404 | 22.0 | 17754 | 2.4148 | 0.6479 |
|
| 76 |
+
| 0.7134 | 23.0 | 18561 | 2.3389 | 0.6761 |
|
| 77 |
+
| 0.6573 | 24.0 | 19368 | 2.6972 | 0.6197 |
|
| 78 |
+
| 0.6848 | 25.0 | 20175 | 2.3375 | 0.6761 |
|
| 79 |
+
| 0.6161 | 26.0 | 20982 | 2.4791 | 0.6620 |
|
| 80 |
+
| 0.6301 | 27.0 | 21789 | 2.3807 | 0.6479 |
|
| 81 |
+
| 0.5758 | 28.0 | 22596 | 2.2243 | 0.6901 |
|
| 82 |
+
| 0.5598 | 29.0 | 23403 | 2.4130 | 0.6620 |
|
| 83 |
+
| 0.6066 | 30.0 | 24210 | 2.4624 | 0.6620 |
|
| 84 |
|
| 85 |
|
| 86 |
### Framework versions
|
| 87 |
|
| 88 |
+
- Transformers 4.44.2
|
| 89 |
+
- Pytorch 2.4.0+cu118
|
| 90 |
+
- Datasets 2.21.0
|
| 91 |
- Tokenizers 0.19.1
|
config.json
CHANGED
|
@@ -118,7 +118,7 @@
|
|
| 118 |
1
|
| 119 |
],
|
| 120 |
"torch_dtype": "float32",
|
| 121 |
-
"transformers_version": "4.
|
| 122 |
"use_weighted_layer_sum": false,
|
| 123 |
"vocab_size": 32,
|
| 124 |
"xvector_output_dim": 512
|
|
|
|
| 118 |
1
|
| 119 |
],
|
| 120 |
"torch_dtype": "float32",
|
| 121 |
+
"transformers_version": "4.44.2",
|
| 122 |
"use_weighted_layer_sum": false,
|
| 123 |
"vocab_size": 32,
|
| 124 |
"xvector_output_dim": 512
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 379887584
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db2434d232d6a9ffd944f00ec930d1eb281de6dd00b09ce923a574e20767d6d6
|
| 3 |
size 379887584
|
trainer_state.json
CHANGED
|
@@ -1,43 +1,507 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 1.0,
|
| 13 |
-
"grad_norm": 5.
|
| 14 |
-
"learning_rate":
|
| 15 |
-
"loss": 1.
|
| 16 |
-
"step":
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 1.0,
|
| 20 |
-
"eval_accuracy": 0.
|
| 21 |
-
"eval_loss": 1.
|
| 22 |
-
"eval_runtime":
|
| 23 |
-
"eval_samples_per_second":
|
| 24 |
-
"eval_steps_per_second":
|
| 25 |
-
"step":
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"epoch":
|
| 29 |
-
"
|
| 30 |
-
"
|
| 31 |
-
"
|
| 32 |
-
"
|
| 33 |
-
|
| 34 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
}
|
| 36 |
],
|
| 37 |
-
"logging_steps":
|
| 38 |
-
"max_steps":
|
| 39 |
"num_input_tokens_seen": 0,
|
| 40 |
-
"num_train_epochs":
|
| 41 |
"save_steps": 500,
|
| 42 |
"stateful_callbacks": {
|
| 43 |
"TrainerControl": {
|
|
@@ -51,7 +515,7 @@
|
|
| 51 |
"attributes": {}
|
| 52 |
}
|
| 53 |
},
|
| 54 |
-
"total_flos":
|
| 55 |
"train_batch_size": 1,
|
| 56 |
"trial_name": null,
|
| 57 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 30.0,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 24210,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 1.0,
|
| 13 |
+
"grad_norm": 5.218414306640625,
|
| 14 |
+
"learning_rate": 9.668731928954978e-06,
|
| 15 |
+
"loss": 1.2041,
|
| 16 |
+
"step": 807
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 1.0,
|
| 20 |
+
"eval_accuracy": 0.3661971688270569,
|
| 21 |
+
"eval_loss": 1.1999133825302124,
|
| 22 |
+
"eval_runtime": 16.8219,
|
| 23 |
+
"eval_samples_per_second": 4.221,
|
| 24 |
+
"eval_steps_per_second": 4.221,
|
| 25 |
+
"step": 807
|
| 26 |
},
|
| 27 |
{
|
| 28 |
+
"epoch": 2.0,
|
| 29 |
+
"grad_norm": 14.095768928527832,
|
| 30 |
+
"learning_rate": 9.335398595621644e-06,
|
| 31 |
+
"loss": 1.067,
|
| 32 |
+
"step": 1614
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 2.0,
|
| 36 |
+
"eval_accuracy": 0.4507042169570923,
|
| 37 |
+
"eval_loss": 1.4613122940063477,
|
| 38 |
+
"eval_runtime": 16.9697,
|
| 39 |
+
"eval_samples_per_second": 4.184,
|
| 40 |
+
"eval_steps_per_second": 4.184,
|
| 41 |
+
"step": 1614
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 3.0,
|
| 45 |
+
"grad_norm": 8.218274116516113,
|
| 46 |
+
"learning_rate": 9.002478314745973e-06,
|
| 47 |
+
"loss": 1.1007,
|
| 48 |
+
"step": 2421
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.0,
|
| 52 |
+
"eval_accuracy": 0.4084506928920746,
|
| 53 |
+
"eval_loss": 1.5213311910629272,
|
| 54 |
+
"eval_runtime": 16.8544,
|
| 55 |
+
"eval_samples_per_second": 4.213,
|
| 56 |
+
"eval_steps_per_second": 4.213,
|
| 57 |
+
"step": 2421
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.0,
|
| 61 |
+
"grad_norm": 43.27208709716797,
|
| 62 |
+
"learning_rate": 8.669558033870302e-06,
|
| 63 |
+
"loss": 1.1945,
|
| 64 |
+
"step": 3228
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 4.0,
|
| 68 |
+
"eval_accuracy": 0.6056337952613831,
|
| 69 |
+
"eval_loss": 1.664249062538147,
|
| 70 |
+
"eval_runtime": 16.9603,
|
| 71 |
+
"eval_samples_per_second": 4.186,
|
| 72 |
+
"eval_steps_per_second": 4.186,
|
| 73 |
+
"step": 3228
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 5.0,
|
| 77 |
+
"grad_norm": 13.959416389465332,
|
| 78 |
+
"learning_rate": 8.336637752994631e-06,
|
| 79 |
+
"loss": 1.3665,
|
| 80 |
+
"step": 4035
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 5.0,
|
| 84 |
+
"eval_accuracy": 0.43661972880363464,
|
| 85 |
+
"eval_loss": 2.9907848834991455,
|
| 86 |
+
"eval_runtime": 16.9865,
|
| 87 |
+
"eval_samples_per_second": 4.18,
|
| 88 |
+
"eval_steps_per_second": 4.18,
|
| 89 |
+
"step": 4035
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.0,
|
| 93 |
+
"grad_norm": 2.819016456604004,
|
| 94 |
+
"learning_rate": 8.003304419661297e-06,
|
| 95 |
+
"loss": 1.4506,
|
| 96 |
+
"step": 4842
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 6.0,
|
| 100 |
+
"eval_accuracy": 0.6056337952613831,
|
| 101 |
+
"eval_loss": 1.9229768514633179,
|
| 102 |
+
"eval_runtime": 17.0241,
|
| 103 |
+
"eval_samples_per_second": 4.171,
|
| 104 |
+
"eval_steps_per_second": 4.171,
|
| 105 |
+
"step": 4842
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.0,
|
| 109 |
+
"grad_norm": 0.291822224855423,
|
| 110 |
+
"learning_rate": 7.669971086327965e-06,
|
| 111 |
+
"loss": 1.495,
|
| 112 |
+
"step": 5649
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 7.0,
|
| 116 |
+
"eval_accuracy": 0.6760563254356384,
|
| 117 |
+
"eval_loss": 1.6813377141952515,
|
| 118 |
+
"eval_runtime": 17.1264,
|
| 119 |
+
"eval_samples_per_second": 4.146,
|
| 120 |
+
"eval_steps_per_second": 4.146,
|
| 121 |
+
"step": 5649
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 8.0,
|
| 125 |
+
"grad_norm": 61.62895584106445,
|
| 126 |
+
"learning_rate": 7.336637752994631e-06,
|
| 127 |
+
"loss": 1.2605,
|
| 128 |
+
"step": 6456
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 8.0,
|
| 132 |
+
"eval_accuracy": 0.6619718074798584,
|
| 133 |
+
"eval_loss": 1.893676519393921,
|
| 134 |
+
"eval_runtime": 17.0263,
|
| 135 |
+
"eval_samples_per_second": 4.17,
|
| 136 |
+
"eval_steps_per_second": 4.17,
|
| 137 |
+
"step": 6456
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 9.0,
|
| 141 |
+
"grad_norm": 137.62139892578125,
|
| 142 |
+
"learning_rate": 7.003717472118959e-06,
|
| 143 |
+
"loss": 1.2713,
|
| 144 |
+
"step": 7263
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.0,
|
| 148 |
+
"eval_accuracy": 0.6901408433914185,
|
| 149 |
+
"eval_loss": 1.6283684968948364,
|
| 150 |
+
"eval_runtime": 17.106,
|
| 151 |
+
"eval_samples_per_second": 4.151,
|
| 152 |
+
"eval_steps_per_second": 4.151,
|
| 153 |
+
"step": 7263
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 10.0,
|
| 157 |
+
"grad_norm": 0.8416326642036438,
|
| 158 |
+
"learning_rate": 6.670384138785626e-06,
|
| 159 |
+
"loss": 1.2608,
|
| 160 |
+
"step": 8070
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 10.0,
|
| 164 |
+
"eval_accuracy": 0.6478873491287231,
|
| 165 |
+
"eval_loss": 1.9437721967697144,
|
| 166 |
+
"eval_runtime": 16.9644,
|
| 167 |
+
"eval_samples_per_second": 4.185,
|
| 168 |
+
"eval_steps_per_second": 4.185,
|
| 169 |
+
"step": 8070
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 11.0,
|
| 173 |
+
"grad_norm": 29.88570785522461,
|
| 174 |
+
"learning_rate": 6.337050805452293e-06,
|
| 175 |
+
"loss": 1.2068,
|
| 176 |
+
"step": 8877
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 11.0,
|
| 180 |
+
"eval_accuracy": 0.7183098793029785,
|
| 181 |
+
"eval_loss": 1.5236804485321045,
|
| 182 |
+
"eval_runtime": 17.1219,
|
| 183 |
+
"eval_samples_per_second": 4.147,
|
| 184 |
+
"eval_steps_per_second": 4.147,
|
| 185 |
+
"step": 8877
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 12.0,
|
| 189 |
+
"grad_norm": 0.15335923433303833,
|
| 190 |
+
"learning_rate": 6.004130524576621e-06,
|
| 191 |
+
"loss": 1.0478,
|
| 192 |
+
"step": 9684
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 12.0,
|
| 196 |
+
"eval_accuracy": 0.6338028311729431,
|
| 197 |
+
"eval_loss": 2.000718832015991,
|
| 198 |
+
"eval_runtime": 16.9949,
|
| 199 |
+
"eval_samples_per_second": 4.178,
|
| 200 |
+
"eval_steps_per_second": 4.178,
|
| 201 |
+
"step": 9684
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 13.0,
|
| 205 |
+
"grad_norm": 0.08654139935970306,
|
| 206 |
+
"learning_rate": 5.670797191243288e-06,
|
| 207 |
+
"loss": 1.1282,
|
| 208 |
+
"step": 10491
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 13.0,
|
| 212 |
+
"eval_accuracy": 0.7464788556098938,
|
| 213 |
+
"eval_loss": 1.5307363271713257,
|
| 214 |
+
"eval_runtime": 17.1388,
|
| 215 |
+
"eval_samples_per_second": 4.143,
|
| 216 |
+
"eval_steps_per_second": 4.143,
|
| 217 |
+
"step": 10491
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 14.0,
|
| 221 |
+
"grad_norm": 598.8184204101562,
|
| 222 |
+
"learning_rate": 5.337463857909955e-06,
|
| 223 |
+
"loss": 0.9433,
|
| 224 |
+
"step": 11298
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 14.0,
|
| 228 |
+
"eval_accuracy": 0.6478873491287231,
|
| 229 |
+
"eval_loss": 2.0042455196380615,
|
| 230 |
+
"eval_runtime": 17.1036,
|
| 231 |
+
"eval_samples_per_second": 4.151,
|
| 232 |
+
"eval_steps_per_second": 4.151,
|
| 233 |
+
"step": 11298
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 15.0,
|
| 237 |
+
"grad_norm": 201.67433166503906,
|
| 238 |
+
"learning_rate": 5.004130524576621e-06,
|
| 239 |
+
"loss": 0.9574,
|
| 240 |
+
"step": 12105
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.0,
|
| 244 |
+
"eval_accuracy": 0.6338028311729431,
|
| 245 |
+
"eval_loss": 2.198476791381836,
|
| 246 |
+
"eval_runtime": 17.0314,
|
| 247 |
+
"eval_samples_per_second": 4.169,
|
| 248 |
+
"eval_steps_per_second": 4.169,
|
| 249 |
+
"step": 12105
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 16.0,
|
| 253 |
+
"grad_norm": 0.1339723765850067,
|
| 254 |
+
"learning_rate": 4.670797191243288e-06,
|
| 255 |
+
"loss": 0.8737,
|
| 256 |
+
"step": 12912
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 16.0,
|
| 260 |
+
"eval_accuracy": 0.6478873491287231,
|
| 261 |
+
"eval_loss": 2.156816005706787,
|
| 262 |
+
"eval_runtime": 17.0458,
|
| 263 |
+
"eval_samples_per_second": 4.165,
|
| 264 |
+
"eval_steps_per_second": 4.165,
|
| 265 |
+
"step": 12912
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 17.0,
|
| 269 |
+
"grad_norm": 2.1163623332977295,
|
| 270 |
+
"learning_rate": 4.338289962825279e-06,
|
| 271 |
+
"loss": 0.8937,
|
| 272 |
+
"step": 13719
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 17.0,
|
| 276 |
+
"eval_accuracy": 0.6197183132171631,
|
| 277 |
+
"eval_loss": 2.2980363368988037,
|
| 278 |
+
"eval_runtime": 16.9937,
|
| 279 |
+
"eval_samples_per_second": 4.178,
|
| 280 |
+
"eval_steps_per_second": 4.178,
|
| 281 |
+
"step": 13719
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 18.0,
|
| 285 |
+
"grad_norm": 542.3316040039062,
|
| 286 |
+
"learning_rate": 4.005369681949608e-06,
|
| 287 |
+
"loss": 0.8681,
|
| 288 |
+
"step": 14526
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 18.0,
|
| 292 |
+
"eval_accuracy": 0.6197183132171631,
|
| 293 |
+
"eval_loss": 2.3267853260040283,
|
| 294 |
+
"eval_runtime": 16.9917,
|
| 295 |
+
"eval_samples_per_second": 4.179,
|
| 296 |
+
"eval_steps_per_second": 4.179,
|
| 297 |
+
"step": 14526
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 19.0,
|
| 301 |
+
"grad_norm": 0.029439380392432213,
|
| 302 |
+
"learning_rate": 3.672036348616275e-06,
|
| 303 |
+
"loss": 0.8005,
|
| 304 |
+
"step": 15333
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 19.0,
|
| 308 |
+
"eval_accuracy": 0.6478873491287231,
|
| 309 |
+
"eval_loss": 2.4827253818511963,
|
| 310 |
+
"eval_runtime": 17.024,
|
| 311 |
+
"eval_samples_per_second": 4.171,
|
| 312 |
+
"eval_steps_per_second": 4.171,
|
| 313 |
+
"step": 15333
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 20.0,
|
| 317 |
+
"grad_norm": 0.3387238681316376,
|
| 318 |
+
"learning_rate": 3.3387030152829415e-06,
|
| 319 |
+
"loss": 0.8176,
|
| 320 |
+
"step": 16140
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 20.0,
|
| 324 |
+
"eval_accuracy": 0.6338028311729431,
|
| 325 |
+
"eval_loss": 2.4842429161071777,
|
| 326 |
+
"eval_runtime": 16.9608,
|
| 327 |
+
"eval_samples_per_second": 4.186,
|
| 328 |
+
"eval_steps_per_second": 4.186,
|
| 329 |
+
"step": 16140
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 21.0,
|
| 333 |
+
"grad_norm": 0.11516053974628448,
|
| 334 |
+
"learning_rate": 3.0053696819496083e-06,
|
| 335 |
+
"loss": 0.8133,
|
| 336 |
+
"step": 16947
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 21.0,
|
| 340 |
+
"eval_accuracy": 0.6760563254356384,
|
| 341 |
+
"eval_loss": 2.061955213546753,
|
| 342 |
+
"eval_runtime": 17.156,
|
| 343 |
+
"eval_samples_per_second": 4.138,
|
| 344 |
+
"eval_steps_per_second": 4.138,
|
| 345 |
+
"step": 16947
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 22.0,
|
| 349 |
+
"grad_norm": 0.07013405114412308,
|
| 350 |
+
"learning_rate": 2.672036348616274e-06,
|
| 351 |
+
"loss": 0.7404,
|
| 352 |
+
"step": 17754
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 22.0,
|
| 356 |
+
"eval_accuracy": 0.6478873491287231,
|
| 357 |
+
"eval_loss": 2.414825916290283,
|
| 358 |
+
"eval_runtime": 17.2261,
|
| 359 |
+
"eval_samples_per_second": 4.122,
|
| 360 |
+
"eval_steps_per_second": 4.122,
|
| 361 |
+
"step": 17754
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 23.0,
|
| 365 |
+
"grad_norm": 0.10270074754953384,
|
| 366 |
+
"learning_rate": 2.3387030152829414e-06,
|
| 367 |
+
"loss": 0.7134,
|
| 368 |
+
"step": 18561
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 23.0,
|
| 372 |
+
"eval_accuracy": 0.6760563254356384,
|
| 373 |
+
"eval_loss": 2.338927984237671,
|
| 374 |
+
"eval_runtime": 17.2024,
|
| 375 |
+
"eval_samples_per_second": 4.127,
|
| 376 |
+
"eval_steps_per_second": 4.127,
|
| 377 |
+
"step": 18561
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 24.0,
|
| 381 |
+
"grad_norm": 633.0289916992188,
|
| 382 |
+
"learning_rate": 2.00578273440727e-06,
|
| 383 |
+
"loss": 0.6573,
|
| 384 |
+
"step": 19368
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 24.0,
|
| 388 |
+
"eval_accuracy": 0.6197183132171631,
|
| 389 |
+
"eval_loss": 2.697195529937744,
|
| 390 |
+
"eval_runtime": 17.2791,
|
| 391 |
+
"eval_samples_per_second": 4.109,
|
| 392 |
+
"eval_steps_per_second": 4.109,
|
| 393 |
+
"step": 19368
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 25.0,
|
| 397 |
+
"grad_norm": 0.12489618360996246,
|
| 398 |
+
"learning_rate": 1.6724494010739365e-06,
|
| 399 |
+
"loss": 0.6848,
|
| 400 |
+
"step": 20175
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 25.0,
|
| 404 |
+
"eval_accuracy": 0.6760563254356384,
|
| 405 |
+
"eval_loss": 2.337477922439575,
|
| 406 |
+
"eval_runtime": 17.079,
|
| 407 |
+
"eval_samples_per_second": 4.157,
|
| 408 |
+
"eval_steps_per_second": 4.157,
|
| 409 |
+
"step": 20175
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 26.0,
|
| 413 |
+
"grad_norm": 0.11721812188625336,
|
| 414 |
+
"learning_rate": 1.3391160677406032e-06,
|
| 415 |
+
"loss": 0.6161,
|
| 416 |
+
"step": 20982
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 26.0,
|
| 420 |
+
"eval_accuracy": 0.6619718074798584,
|
| 421 |
+
"eval_loss": 2.479071617126465,
|
| 422 |
+
"eval_runtime": 17.1585,
|
| 423 |
+
"eval_samples_per_second": 4.138,
|
| 424 |
+
"eval_steps_per_second": 4.138,
|
| 425 |
+
"step": 20982
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 27.0,
|
| 429 |
+
"grad_norm": 0.09104170650243759,
|
| 430 |
+
"learning_rate": 1.0057827344072698e-06,
|
| 431 |
+
"loss": 0.6301,
|
| 432 |
+
"step": 21789
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 27.0,
|
| 436 |
+
"eval_accuracy": 0.6478873491287231,
|
| 437 |
+
"eval_loss": 2.3807287216186523,
|
| 438 |
+
"eval_runtime": 17.3153,
|
| 439 |
+
"eval_samples_per_second": 4.1,
|
| 440 |
+
"eval_steps_per_second": 4.1,
|
| 441 |
+
"step": 21789
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 28.0,
|
| 445 |
+
"grad_norm": 0.048452723771333694,
|
| 446 |
+
"learning_rate": 6.724494010739364e-07,
|
| 447 |
+
"loss": 0.5758,
|
| 448 |
+
"step": 22596
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 28.0,
|
| 452 |
+
"eval_accuracy": 0.6901408433914185,
|
| 453 |
+
"eval_loss": 2.224256753921509,
|
| 454 |
+
"eval_runtime": 17.1054,
|
| 455 |
+
"eval_samples_per_second": 4.151,
|
| 456 |
+
"eval_steps_per_second": 4.151,
|
| 457 |
+
"step": 22596
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 29.0,
|
| 461 |
+
"grad_norm": 79.71731567382812,
|
| 462 |
+
"learning_rate": 3.395291201982652e-07,
|
| 463 |
+
"loss": 0.5598,
|
| 464 |
+
"step": 23403
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 29.0,
|
| 468 |
+
"eval_accuracy": 0.6619718074798584,
|
| 469 |
+
"eval_loss": 2.41304874420166,
|
| 470 |
+
"eval_runtime": 17.2038,
|
| 471 |
+
"eval_samples_per_second": 4.127,
|
| 472 |
+
"eval_steps_per_second": 4.127,
|
| 473 |
+
"step": 23403
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 30.0,
|
| 477 |
+
"grad_norm": 0.05571676418185234,
|
| 478 |
+
"learning_rate": 7.021891780256092e-09,
|
| 479 |
+
"loss": 0.6066,
|
| 480 |
+
"step": 24210
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 30.0,
|
| 484 |
+
"eval_accuracy": 0.6619718074798584,
|
| 485 |
+
"eval_loss": 2.4623985290527344,
|
| 486 |
+
"eval_runtime": 17.2965,
|
| 487 |
+
"eval_samples_per_second": 4.105,
|
| 488 |
+
"eval_steps_per_second": 4.105,
|
| 489 |
+
"step": 24210
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 30.0,
|
| 493 |
+
"step": 24210,
|
| 494 |
+
"total_flos": 6.696180305029608e+18,
|
| 495 |
+
"train_loss": 0.9601948892119895,
|
| 496 |
+
"train_runtime": 9628.1801,
|
| 497 |
+
"train_samples_per_second": 2.514,
|
| 498 |
+
"train_steps_per_second": 2.514
|
| 499 |
}
|
| 500 |
],
|
| 501 |
+
"logging_steps": 12,
|
| 502 |
+
"max_steps": 24210,
|
| 503 |
"num_input_tokens_seen": 0,
|
| 504 |
+
"num_train_epochs": 30,
|
| 505 |
"save_steps": 500,
|
| 506 |
"stateful_callbacks": {
|
| 507 |
"TrainerControl": {
|
|
|
|
| 515 |
"attributes": {}
|
| 516 |
}
|
| 517 |
},
|
| 518 |
+
"total_flos": 6.696180305029608e+18,
|
| 519 |
"train_batch_size": 1,
|
| 520 |
"trial_name": null,
|
| 521 |
"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
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be9bdbc8f61337ff7a8ee5b484d68488f187a05ca25bcba8668068a6feeee799
|
| 3 |
+
size 5240
|