Upload 16 files
Browse files- README.md +38 -22
- all_results.json +10 -10
- config.json +1 -1
- eval_results.json +7 -7
- runs/Mar15_17-26-14_srvgpu/1678901187.8142433/events.out.tfevents.1678901187.srvgpu.2936559.1 +3 -0
- runs/Mar15_17-26-14_srvgpu/events.out.tfevents.1678901187.srvgpu.2936559.0 +3 -0
- runs/Mar15_17-26-14_srvgpu/events.out.tfevents.1678911557.srvgpu.2936559.2 +3 -0
- tokenizer_config.json +1 -1
- train_results.json +4 -4
- trainer_state.json +291 -105
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -5,19 +5,19 @@ tags:
|
|
| 5 |
metrics:
|
| 6 |
- accuracy
|
| 7 |
model-index:
|
| 8 |
-
- name:
|
| 9 |
results: []
|
| 10 |
---
|
| 11 |
|
| 12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
|
| 15 |
-
#
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [avuhong/ParvoGPT2](https://huggingface.co/avuhong/ParvoGPT2) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
-
- Accuracy: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
@@ -36,7 +36,7 @@ More information needed
|
|
| 36 |
### Training hyperparameters
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
-
- learning_rate:
|
| 40 |
- train_batch_size: 1
|
| 41 |
- eval_batch_size: 1
|
| 42 |
- seed: 42
|
|
@@ -47,29 +47,45 @@ The following hyperparameters were used during training:
|
|
| 47 |
- total_eval_batch_size: 2
|
| 48 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
- lr_scheduler_type: linear
|
| 50 |
-
- num_epochs:
|
| 51 |
- mixed_precision_training: Native AMP
|
| 52 |
|
| 53 |
### Training results
|
| 54 |
|
| 55 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 56 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 57 |
-
| No log | 1.0 | 220 | 1.
|
| 58 |
-
| No log | 2.0 | 440 |
|
| 59 |
-
| 1.
|
| 60 |
-
| 1.
|
| 61 |
-
|
|
| 62 |
-
|
|
| 63 |
-
|
|
| 64 |
-
|
|
| 65 |
-
|
|
| 66 |
-
|
|
| 67 |
-
|
|
| 68 |
-
|
|
| 69 |
-
|
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
|
| 74 |
|
| 75 |
### Framework versions
|
|
|
|
| 5 |
metrics:
|
| 6 |
- accuracy
|
| 7 |
model-index:
|
| 8 |
+
- name: output_v3
|
| 9 |
results: []
|
| 10 |
---
|
| 11 |
|
| 12 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
|
| 15 |
+
# output_v3
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [avuhong/ParvoGPT2](https://huggingface.co/avuhong/ParvoGPT2) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.4775
|
| 20 |
+
- Accuracy: 0.9290
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 36 |
### Training hyperparameters
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 5e-05
|
| 40 |
- train_batch_size: 1
|
| 41 |
- eval_batch_size: 1
|
| 42 |
- seed: 42
|
|
|
|
| 47 |
- total_eval_batch_size: 2
|
| 48 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
- lr_scheduler_type: linear
|
| 50 |
+
- num_epochs: 32.0
|
| 51 |
- mixed_precision_training: Native AMP
|
| 52 |
|
| 53 |
### Training results
|
| 54 |
|
| 55 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 56 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 57 |
+
| No log | 1.0 | 220 | 1.1623 | 0.8225 |
|
| 58 |
+
| No log | 2.0 | 440 | 0.9566 | 0.8539 |
|
| 59 |
+
| 1.1942 | 3.0 | 660 | 0.8456 | 0.8709 |
|
| 60 |
+
| 1.1942 | 4.0 | 880 | 0.7719 | 0.8801 |
|
| 61 |
+
| 0.7805 | 5.0 | 1100 | 0.7224 | 0.8872 |
|
| 62 |
+
| 0.7805 | 6.0 | 1320 | 0.6895 | 0.8928 |
|
| 63 |
+
| 0.6257 | 7.0 | 1540 | 0.6574 | 0.8972 |
|
| 64 |
+
| 0.6257 | 8.0 | 1760 | 0.6289 | 0.9014 |
|
| 65 |
+
| 0.6257 | 9.0 | 1980 | 0.6054 | 0.9045 |
|
| 66 |
+
| 0.5385 | 10.0 | 2200 | 0.5881 | 0.9077 |
|
| 67 |
+
| 0.5385 | 11.0 | 2420 | 0.5709 | 0.9102 |
|
| 68 |
+
| 0.4778 | 12.0 | 2640 | 0.5591 | 0.9121 |
|
| 69 |
+
| 0.4778 | 13.0 | 2860 | 0.5497 | 0.9143 |
|
| 70 |
+
| 0.427 | 14.0 | 3080 | 0.5385 | 0.9161 |
|
| 71 |
+
| 0.427 | 15.0 | 3300 | 0.5258 | 0.9180 |
|
| 72 |
+
| 0.394 | 16.0 | 3520 | 0.5170 | 0.9195 |
|
| 73 |
+
| 0.394 | 17.0 | 3740 | 0.5157 | 0.9212 |
|
| 74 |
+
| 0.394 | 18.0 | 3960 | 0.5038 | 0.9221 |
|
| 75 |
+
| 0.363 | 19.0 | 4180 | 0.4977 | 0.9234 |
|
| 76 |
+
| 0.363 | 20.0 | 4400 | 0.4976 | 0.9236 |
|
| 77 |
+
| 0.3392 | 21.0 | 4620 | 0.4924 | 0.9247 |
|
| 78 |
+
| 0.3392 | 22.0 | 4840 | 0.4888 | 0.9255 |
|
| 79 |
+
| 0.33 | 23.0 | 5060 | 0.4890 | 0.9262 |
|
| 80 |
+
| 0.33 | 24.0 | 5280 | 0.4856 | 0.9268 |
|
| 81 |
+
| 0.3058 | 25.0 | 5500 | 0.4803 | 0.9275 |
|
| 82 |
+
| 0.3058 | 26.0 | 5720 | 0.4785 | 0.9277 |
|
| 83 |
+
| 0.3058 | 27.0 | 5940 | 0.4813 | 0.9281 |
|
| 84 |
+
| 0.2973 | 28.0 | 6160 | 0.4799 | 0.9282 |
|
| 85 |
+
| 0.2973 | 29.0 | 6380 | 0.4773 | 0.9285 |
|
| 86 |
+
| 0.2931 | 30.0 | 6600 | 0.4778 | 0.9286 |
|
| 87 |
+
| 0.2931 | 31.0 | 6820 | 0.4756 | 0.9290 |
|
| 88 |
+
| 0.2879 | 32.0 | 7040 | 0.4775 | 0.9290 |
|
| 89 |
|
| 90 |
|
| 91 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
"eval_samples": 91,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
-
"perplexity":
|
| 10 |
-
"train_loss":
|
| 11 |
-
"train_runtime":
|
| 12 |
"train_samples": 1762,
|
| 13 |
"train_samples_per_second": 5.442,
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 32.0,
|
| 3 |
+
"eval_accuracy": 0.9289527676624451,
|
| 4 |
+
"eval_loss": 0.47753414511680603,
|
| 5 |
+
"eval_runtime": 6.0712,
|
| 6 |
"eval_samples": 91,
|
| 7 |
+
"eval_samples_per_second": 14.989,
|
| 8 |
+
"eval_steps_per_second": 7.577,
|
| 9 |
+
"perplexity": 1.6120943064226643,
|
| 10 |
+
"train_loss": 0.47427067851478405,
|
| 11 |
+
"train_runtime": 10360.152,
|
| 12 |
"train_samples": 1762,
|
| 13 |
"train_samples_per_second": 5.442,
|
| 14 |
+
"train_steps_per_second": 0.68
|
| 15 |
}
|
config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"architectures": [
|
| 5 |
"GPT2LMHeadModel"
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "avuhong/ParvoGPT2",
|
| 3 |
"activation_function": "gelu_new",
|
| 4 |
"architectures": [
|
| 5 |
"GPT2LMHeadModel"
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
"eval_samples": 91,
|
| 7 |
-
"eval_samples_per_second":
|
| 8 |
-
"eval_steps_per_second":
|
| 9 |
-
"perplexity":
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 32.0,
|
| 3 |
+
"eval_accuracy": 0.9289527676624451,
|
| 4 |
+
"eval_loss": 0.47753414511680603,
|
| 5 |
+
"eval_runtime": 6.0712,
|
| 6 |
"eval_samples": 91,
|
| 7 |
+
"eval_samples_per_second": 14.989,
|
| 8 |
+
"eval_steps_per_second": 7.577,
|
| 9 |
+
"perplexity": 1.6120943064226643
|
| 10 |
}
|
runs/Mar15_17-26-14_srvgpu/1678901187.8142433/events.out.tfevents.1678901187.srvgpu.2936559.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:115496e7d79030d313da0cb77622a6aa7a29c7c51cc9c631d90574c2ad318d67
|
| 3 |
+
size 5643
|
runs/Mar15_17-26-14_srvgpu/events.out.tfevents.1678901187.srvgpu.2936559.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1fdcee8d6bf9c043faf357536cbe68e370e845f5a47aa1679291e96c05b51d5d
|
| 3 |
+
size 16954
|
runs/Mar15_17-26-14_srvgpu/events.out.tfevents.1678911557.srvgpu.2936559.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5bc6a5fd4556ffd4c7220a94911b8d060aa146107d8ab1774b9a6c3c343b2534
|
| 3 |
+
size 363
|
tokenizer_config.json
CHANGED
|
@@ -3,7 +3,7 @@
|
|
| 3 |
"bos_token": "<|endoftext|>",
|
| 4 |
"eos_token": "<|endoftext|>",
|
| 5 |
"model_max_length": 1000000000000000019884624838656,
|
| 6 |
-
"name_or_path": "
|
| 7 |
"special_tokens_map_file": "/home/avuhong/.cache/huggingface/hub/models--nferruz--ProtGPT2/snapshots/afbc64cbb1e9c2b11f66f1f7c66d0be28df32b1b/special_tokens_map.json",
|
| 8 |
"tokenizer_class": "GPT2Tokenizer",
|
| 9 |
"unk_token": "<|endoftext|>"
|
|
|
|
| 3 |
"bos_token": "<|endoftext|>",
|
| 4 |
"eos_token": "<|endoftext|>",
|
| 5 |
"model_max_length": 1000000000000000019884624838656,
|
| 6 |
+
"name_or_path": "avuhong/ParvoGPT2",
|
| 7 |
"special_tokens_map_file": "/home/avuhong/.cache/huggingface/hub/models--nferruz--ProtGPT2/snapshots/afbc64cbb1e9c2b11f66f1f7c66d0be28df32b1b/special_tokens_map.json",
|
| 8 |
"tokenizer_class": "GPT2Tokenizer",
|
| 9 |
"unk_token": "<|endoftext|>"
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 1762,
|
| 6 |
"train_samples_per_second": 5.442,
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 32.0,
|
| 3 |
+
"train_loss": 0.47427067851478405,
|
| 4 |
+
"train_runtime": 10360.152,
|
| 5 |
"train_samples": 1762,
|
| 6 |
"train_samples_per_second": 5.442,
|
| 7 |
+
"train_steps_per_second": 0.68
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,211 +1,397 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch":
|
| 5 |
-
"global_step":
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 1.0,
|
| 12 |
-
"eval_accuracy": 0.
|
| 13 |
-
"eval_loss": 1.
|
| 14 |
-
"eval_runtime": 5.
|
| 15 |
-
"eval_samples_per_second": 15.
|
| 16 |
-
"eval_steps_per_second": 7.
|
| 17 |
"step": 220
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 2.0,
|
| 21 |
-
"eval_accuracy": 0.
|
| 22 |
-
"eval_loss":
|
| 23 |
-
"eval_runtime": 5.
|
| 24 |
-
"eval_samples_per_second": 15.
|
| 25 |
-
"eval_steps_per_second": 7.
|
| 26 |
"step": 440
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 2.27,
|
| 30 |
-
"learning_rate":
|
| 31 |
-
"loss": 1.
|
| 32 |
"step": 500
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 3.0,
|
| 36 |
-
"eval_accuracy": 0.
|
| 37 |
-
"eval_loss":
|
| 38 |
-
"eval_runtime": 5.
|
| 39 |
-
"eval_samples_per_second": 15.
|
| 40 |
-
"eval_steps_per_second": 7.
|
| 41 |
"step": 660
|
| 42 |
},
|
| 43 |
{
|
| 44 |
"epoch": 4.0,
|
| 45 |
-
"eval_accuracy": 0.
|
| 46 |
-
"eval_loss":
|
| 47 |
-
"eval_runtime": 5.
|
| 48 |
-
"eval_samples_per_second": 16.
|
| 49 |
-
"eval_steps_per_second": 8.
|
| 50 |
"step": 880
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 4.54,
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss":
|
| 56 |
"step": 1000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 5.0,
|
| 60 |
-
"eval_accuracy": 0.
|
| 61 |
-
"eval_loss":
|
| 62 |
-
"eval_runtime": 5.
|
| 63 |
-
"eval_samples_per_second": 15.
|
| 64 |
-
"eval_steps_per_second": 7.
|
| 65 |
"step": 1100
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 6.0,
|
| 69 |
-
"eval_accuracy": 0.
|
| 70 |
-
"eval_loss":
|
| 71 |
-
"eval_runtime": 5.
|
| 72 |
-
"eval_samples_per_second": 15.
|
| 73 |
-
"eval_steps_per_second": 7.
|
| 74 |
"step": 1320
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 6.82,
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss":
|
| 80 |
"step": 1500
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 7.0,
|
| 84 |
-
"eval_accuracy": 0.
|
| 85 |
-
"eval_loss":
|
| 86 |
-
"eval_runtime": 5.
|
| 87 |
-
"eval_samples_per_second": 15.
|
| 88 |
-
"eval_steps_per_second": 7.
|
| 89 |
"step": 1540
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 8.0,
|
| 93 |
-
"eval_accuracy": 0.
|
| 94 |
-
"eval_loss":
|
| 95 |
-
"eval_runtime": 5.
|
| 96 |
-
"eval_samples_per_second": 15.
|
| 97 |
-
"eval_steps_per_second":
|
| 98 |
"step": 1760
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 9.0,
|
| 102 |
-
"eval_accuracy": 0.
|
| 103 |
-
"eval_loss":
|
| 104 |
-
"eval_runtime": 5.
|
| 105 |
-
"eval_samples_per_second":
|
| 106 |
-
"eval_steps_per_second":
|
| 107 |
"step": 1980
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 9.09,
|
| 111 |
-
"learning_rate":
|
| 112 |
-
"loss":
|
| 113 |
"step": 2000
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 10.0,
|
| 117 |
-
"eval_accuracy": 0.
|
| 118 |
-
"eval_loss":
|
| 119 |
-
"eval_runtime":
|
| 120 |
-
"eval_samples_per_second":
|
| 121 |
-
"eval_steps_per_second":
|
| 122 |
"step": 2200
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 11.0,
|
| 126 |
-
"eval_accuracy": 0.
|
| 127 |
-
"eval_loss":
|
| 128 |
-
"eval_runtime": 5.
|
| 129 |
-
"eval_samples_per_second": 15.
|
| 130 |
-
"eval_steps_per_second": 7.
|
| 131 |
"step": 2420
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 11.36,
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss":
|
| 137 |
"step": 2500
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 12.0,
|
| 141 |
-
"eval_accuracy": 0.
|
| 142 |
-
"eval_loss":
|
| 143 |
-
"eval_runtime": 5.
|
| 144 |
-
"eval_samples_per_second": 15.
|
| 145 |
-
"eval_steps_per_second": 7.
|
| 146 |
"step": 2640
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 13.0,
|
| 150 |
-
"eval_accuracy": 0.
|
| 151 |
-
"eval_loss": 0.
|
| 152 |
-
"eval_runtime": 5.
|
| 153 |
-
"eval_samples_per_second": 15.
|
| 154 |
-
"eval_steps_per_second": 7.
|
| 155 |
"step": 2860
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 13.64,
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss": 0.
|
| 161 |
"step": 3000
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 14.0,
|
| 165 |
-
"eval_accuracy": 0.
|
| 166 |
-
"eval_loss": 0.
|
| 167 |
-
"eval_runtime": 5.
|
| 168 |
-
"eval_samples_per_second": 15.
|
| 169 |
-
"eval_steps_per_second": 7.
|
| 170 |
"step": 3080
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 15.0,
|
| 174 |
-
"eval_accuracy": 0.
|
| 175 |
-
"eval_loss": 0.
|
| 176 |
-
"eval_runtime": 5.
|
| 177 |
-
"eval_samples_per_second": 15.
|
| 178 |
-
"eval_steps_per_second": 7.
|
| 179 |
"step": 3300
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 15.91,
|
| 183 |
-
"learning_rate":
|
| 184 |
-
"loss": 0.
|
| 185 |
"step": 3500
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 16.0,
|
| 189 |
-
"eval_accuracy": 0.
|
| 190 |
-
"eval_loss": 0.
|
| 191 |
-
"eval_runtime": 5.
|
| 192 |
-
"eval_samples_per_second":
|
| 193 |
-
"eval_steps_per_second":
|
| 194 |
"step": 3520
|
| 195 |
},
|
| 196 |
{
|
| 197 |
-
"epoch":
|
| 198 |
-
"
|
| 199 |
-
"
|
| 200 |
-
"
|
| 201 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 202 |
"train_samples_per_second": 5.442,
|
| 203 |
-
"train_steps_per_second": 0.
|
| 204 |
}
|
| 205 |
],
|
| 206 |
-
"max_steps":
|
| 207 |
-
"num_train_epochs":
|
| 208 |
-
"total_flos":
|
| 209 |
"trial_name": null,
|
| 210 |
"trial_params": null
|
| 211 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 31.998864926220204,
|
| 5 |
+
"global_step": 7040,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
"epoch": 1.0,
|
| 12 |
+
"eval_accuracy": 0.8224893386183709,
|
| 13 |
+
"eval_loss": 1.1622651815414429,
|
| 14 |
+
"eval_runtime": 5.9816,
|
| 15 |
+
"eval_samples_per_second": 15.213,
|
| 16 |
+
"eval_steps_per_second": 7.69,
|
| 17 |
"step": 220
|
| 18 |
},
|
| 19 |
{
|
| 20 |
"epoch": 2.0,
|
| 21 |
+
"eval_accuracy": 0.8538665635439829,
|
| 22 |
+
"eval_loss": 0.9566460251808167,
|
| 23 |
+
"eval_runtime": 5.9768,
|
| 24 |
+
"eval_samples_per_second": 15.226,
|
| 25 |
+
"eval_steps_per_second": 7.696,
|
| 26 |
"step": 440
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 2.27,
|
| 30 |
+
"learning_rate": 4.6448863636363636e-05,
|
| 31 |
+
"loss": 1.1942,
|
| 32 |
"step": 500
|
| 33 |
},
|
| 34 |
{
|
| 35 |
"epoch": 3.0,
|
| 36 |
+
"eval_accuracy": 0.8709140322043548,
|
| 37 |
+
"eval_loss": 0.8456286191940308,
|
| 38 |
+
"eval_runtime": 5.9871,
|
| 39 |
+
"eval_samples_per_second": 15.199,
|
| 40 |
+
"eval_steps_per_second": 7.683,
|
| 41 |
"step": 660
|
| 42 |
},
|
| 43 |
{
|
| 44 |
"epoch": 4.0,
|
| 45 |
+
"eval_accuracy": 0.8801198801198801,
|
| 46 |
+
"eval_loss": 0.7718582153320312,
|
| 47 |
+
"eval_runtime": 5.6728,
|
| 48 |
+
"eval_samples_per_second": 16.041,
|
| 49 |
+
"eval_steps_per_second": 8.109,
|
| 50 |
"step": 880
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 4.54,
|
| 54 |
+
"learning_rate": 4.289772727272727e-05,
|
| 55 |
+
"loss": 0.7805,
|
| 56 |
"step": 1000
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 5.0,
|
| 60 |
+
"eval_accuracy": 0.8871880807364678,
|
| 61 |
+
"eval_loss": 0.7224407196044922,
|
| 62 |
+
"eval_runtime": 5.988,
|
| 63 |
+
"eval_samples_per_second": 15.197,
|
| 64 |
+
"eval_steps_per_second": 7.682,
|
| 65 |
"step": 1100
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 6.0,
|
| 69 |
+
"eval_accuracy": 0.892816860558796,
|
| 70 |
+
"eval_loss": 0.6894707679748535,
|
| 71 |
+
"eval_runtime": 5.9958,
|
| 72 |
+
"eval_samples_per_second": 15.177,
|
| 73 |
+
"eval_steps_per_second": 7.672,
|
| 74 |
"step": 1320
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 6.82,
|
| 78 |
+
"learning_rate": 3.934659090909091e-05,
|
| 79 |
+
"loss": 0.6257,
|
| 80 |
"step": 1500
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 7.0,
|
| 84 |
+
"eval_accuracy": 0.8972103165651553,
|
| 85 |
+
"eval_loss": 0.6574080586433411,
|
| 86 |
+
"eval_runtime": 5.9909,
|
| 87 |
+
"eval_samples_per_second": 15.19,
|
| 88 |
+
"eval_steps_per_second": 7.678,
|
| 89 |
"step": 1540
|
| 90 |
},
|
| 91 |
{
|
| 92 |
"epoch": 8.0,
|
| 93 |
+
"eval_accuracy": 0.9014426433781273,
|
| 94 |
+
"eval_loss": 0.6289474368095398,
|
| 95 |
+
"eval_runtime": 5.6911,
|
| 96 |
+
"eval_samples_per_second": 15.99,
|
| 97 |
+
"eval_steps_per_second": 8.083,
|
| 98 |
"step": 1760
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 9.0,
|
| 102 |
+
"eval_accuracy": 0.9045470658373884,
|
| 103 |
+
"eval_loss": 0.6054205298423767,
|
| 104 |
+
"eval_runtime": 5.6786,
|
| 105 |
+
"eval_samples_per_second": 16.025,
|
| 106 |
+
"eval_steps_per_second": 8.101,
|
| 107 |
"step": 1980
|
| 108 |
},
|
| 109 |
{
|
| 110 |
"epoch": 9.09,
|
| 111 |
+
"learning_rate": 3.579545454545455e-05,
|
| 112 |
+
"loss": 0.5385,
|
| 113 |
"step": 2000
|
| 114 |
},
|
| 115 |
{
|
| 116 |
"epoch": 10.0,
|
| 117 |
+
"eval_accuracy": 0.9076622302428754,
|
| 118 |
+
"eval_loss": 0.5881273746490479,
|
| 119 |
+
"eval_runtime": 6.0014,
|
| 120 |
+
"eval_samples_per_second": 15.163,
|
| 121 |
+
"eval_steps_per_second": 7.665,
|
| 122 |
"step": 2200
|
| 123 |
},
|
| 124 |
{
|
| 125 |
"epoch": 11.0,
|
| 126 |
+
"eval_accuracy": 0.9101865876059424,
|
| 127 |
+
"eval_loss": 0.5709272623062134,
|
| 128 |
+
"eval_runtime": 5.9886,
|
| 129 |
+
"eval_samples_per_second": 15.195,
|
| 130 |
+
"eval_steps_per_second": 7.681,
|
| 131 |
"step": 2420
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 11.36,
|
| 135 |
+
"learning_rate": 3.2244318181818185e-05,
|
| 136 |
+
"loss": 0.4778,
|
| 137 |
"step": 2500
|
| 138 |
},
|
| 139 |
{
|
| 140 |
"epoch": 12.0,
|
| 141 |
+
"eval_accuracy": 0.9120986540341379,
|
| 142 |
+
"eval_loss": 0.5591339468955994,
|
| 143 |
+
"eval_runtime": 5.9874,
|
| 144 |
+
"eval_samples_per_second": 15.199,
|
| 145 |
+
"eval_steps_per_second": 7.683,
|
| 146 |
"step": 2640
|
| 147 |
},
|
| 148 |
{
|
| 149 |
"epoch": 13.0,
|
| 150 |
+
"eval_accuracy": 0.9142900110642046,
|
| 151 |
+
"eval_loss": 0.5496613383293152,
|
| 152 |
+
"eval_runtime": 5.9826,
|
| 153 |
+
"eval_samples_per_second": 15.211,
|
| 154 |
+
"eval_steps_per_second": 7.689,
|
| 155 |
"step": 2860
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 13.64,
|
| 159 |
+
"learning_rate": 2.870028409090909e-05,
|
| 160 |
+
"loss": 0.427,
|
| 161 |
"step": 3000
|
| 162 |
},
|
| 163 |
{
|
| 164 |
"epoch": 14.0,
|
| 165 |
+
"eval_accuracy": 0.9161053999763678,
|
| 166 |
+
"eval_loss": 0.5385328531265259,
|
| 167 |
+
"eval_runtime": 5.9885,
|
| 168 |
+
"eval_samples_per_second": 15.196,
|
| 169 |
+
"eval_steps_per_second": 7.681,
|
| 170 |
"step": 3080
|
| 171 |
},
|
| 172 |
{
|
| 173 |
"epoch": 15.0,
|
| 174 |
+
"eval_accuracy": 0.9179637566734341,
|
| 175 |
+
"eval_loss": 0.5258467793464661,
|
| 176 |
+
"eval_runtime": 5.9812,
|
| 177 |
+
"eval_samples_per_second": 15.214,
|
| 178 |
+
"eval_steps_per_second": 7.691,
|
| 179 |
"step": 3300
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 15.91,
|
| 183 |
+
"learning_rate": 2.5149147727272725e-05,
|
| 184 |
+
"loss": 0.394,
|
| 185 |
"step": 3500
|
| 186 |
},
|
| 187 |
{
|
| 188 |
"epoch": 16.0,
|
| 189 |
+
"eval_accuracy": 0.9195428227686292,
|
| 190 |
+
"eval_loss": 0.5170450806617737,
|
| 191 |
+
"eval_runtime": 5.6775,
|
| 192 |
+
"eval_samples_per_second": 16.028,
|
| 193 |
+
"eval_steps_per_second": 8.102,
|
| 194 |
"step": 3520
|
| 195 |
},
|
| 196 |
{
|
| 197 |
+
"epoch": 17.0,
|
| 198 |
+
"eval_accuracy": 0.9211648566487276,
|
| 199 |
+
"eval_loss": 0.5156892538070679,
|
| 200 |
+
"eval_runtime": 5.9945,
|
| 201 |
+
"eval_samples_per_second": 15.18,
|
| 202 |
+
"eval_steps_per_second": 7.674,
|
| 203 |
+
"step": 3740
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 18.0,
|
| 207 |
+
"eval_accuracy": 0.9220671801316963,
|
| 208 |
+
"eval_loss": 0.5037761926651001,
|
| 209 |
+
"eval_runtime": 5.687,
|
| 210 |
+
"eval_samples_per_second": 16.001,
|
| 211 |
+
"eval_steps_per_second": 8.089,
|
| 212 |
+
"step": 3960
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 18.18,
|
| 216 |
+
"learning_rate": 2.1598011363636363e-05,
|
| 217 |
+
"loss": 0.363,
|
| 218 |
+
"step": 4000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 19.0,
|
| 222 |
+
"eval_accuracy": 0.9233884395174717,
|
| 223 |
+
"eval_loss": 0.49766021966934204,
|
| 224 |
+
"eval_runtime": 5.6849,
|
| 225 |
+
"eval_samples_per_second": 16.007,
|
| 226 |
+
"eval_steps_per_second": 8.092,
|
| 227 |
+
"step": 4180
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 20.0,
|
| 231 |
+
"eval_accuracy": 0.9236462462268914,
|
| 232 |
+
"eval_loss": 0.4975946843624115,
|
| 233 |
+
"eval_runtime": 5.9923,
|
| 234 |
+
"eval_samples_per_second": 15.186,
|
| 235 |
+
"eval_steps_per_second": 7.677,
|
| 236 |
+
"step": 4400
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 20.45,
|
| 240 |
+
"learning_rate": 1.8046875000000003e-05,
|
| 241 |
+
"loss": 0.3392,
|
| 242 |
+
"step": 4500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 21.0,
|
| 246 |
+
"eval_accuracy": 0.9246882150107957,
|
| 247 |
+
"eval_loss": 0.49241966009140015,
|
| 248 |
+
"eval_runtime": 5.9923,
|
| 249 |
+
"eval_samples_per_second": 15.186,
|
| 250 |
+
"eval_steps_per_second": 7.677,
|
| 251 |
+
"step": 4620
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 22.0,
|
| 255 |
+
"eval_accuracy": 0.9255046029239578,
|
| 256 |
+
"eval_loss": 0.4887617826461792,
|
| 257 |
+
"eval_runtime": 5.69,
|
| 258 |
+
"eval_samples_per_second": 15.993,
|
| 259 |
+
"eval_steps_per_second": 8.084,
|
| 260 |
+
"step": 4840
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 22.73,
|
| 264 |
+
"learning_rate": 1.4495738636363637e-05,
|
| 265 |
+
"loss": 0.33,
|
| 266 |
+
"step": 5000
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 23.0,
|
| 270 |
+
"eval_accuracy": 0.9262028294286359,
|
| 271 |
+
"eval_loss": 0.4889785051345825,
|
| 272 |
+
"eval_runtime": 6.0024,
|
| 273 |
+
"eval_samples_per_second": 15.161,
|
| 274 |
+
"eval_steps_per_second": 7.664,
|
| 275 |
+
"step": 5060
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 24.0,
|
| 279 |
+
"eval_accuracy": 0.9267936364710558,
|
| 280 |
+
"eval_loss": 0.4856303334236145,
|
| 281 |
+
"eval_runtime": 5.9867,
|
| 282 |
+
"eval_samples_per_second": 15.2,
|
| 283 |
+
"eval_steps_per_second": 7.684,
|
| 284 |
+
"step": 5280
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 25.0,
|
| 288 |
+
"learning_rate": 1.0951704545454545e-05,
|
| 289 |
+
"loss": 0.3058,
|
| 290 |
+
"step": 5500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 25.0,
|
| 294 |
+
"eval_accuracy": 0.9275348307606373,
|
| 295 |
+
"eval_loss": 0.4802783131599426,
|
| 296 |
+
"eval_runtime": 5.6869,
|
| 297 |
+
"eval_samples_per_second": 16.002,
|
| 298 |
+
"eval_steps_per_second": 8.089,
|
| 299 |
+
"step": 5500
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 26.0,
|
| 303 |
+
"eval_accuracy": 0.9277389277389277,
|
| 304 |
+
"eval_loss": 0.47845765948295593,
|
| 305 |
+
"eval_runtime": 5.9773,
|
| 306 |
+
"eval_samples_per_second": 15.224,
|
| 307 |
+
"eval_steps_per_second": 7.696,
|
| 308 |
+
"step": 5720
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 27.0,
|
| 312 |
+
"eval_accuracy": 0.9280611861257022,
|
| 313 |
+
"eval_loss": 0.4813348948955536,
|
| 314 |
+
"eval_runtime": 5.6869,
|
| 315 |
+
"eval_samples_per_second": 16.002,
|
| 316 |
+
"eval_steps_per_second": 8.089,
|
| 317 |
+
"step": 5940
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 27.27,
|
| 321 |
+
"learning_rate": 7.407670454545455e-06,
|
| 322 |
+
"loss": 0.2973,
|
| 323 |
+
"step": 6000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 28.0,
|
| 327 |
+
"eval_accuracy": 0.9281793475341863,
|
| 328 |
+
"eval_loss": 0.4798637628555298,
|
| 329 |
+
"eval_runtime": 5.6812,
|
| 330 |
+
"eval_samples_per_second": 16.018,
|
| 331 |
+
"eval_steps_per_second": 8.097,
|
| 332 |
+
"step": 6160
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 29.0,
|
| 336 |
+
"eval_accuracy": 0.9285230898134124,
|
| 337 |
+
"eval_loss": 0.47730037569999695,
|
| 338 |
+
"eval_runtime": 5.9912,
|
| 339 |
+
"eval_samples_per_second": 15.189,
|
| 340 |
+
"eval_steps_per_second": 7.678,
|
| 341 |
+
"step": 6380
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 29.54,
|
| 345 |
+
"learning_rate": 3.856534090909091e-06,
|
| 346 |
+
"loss": 0.2931,
|
| 347 |
+
"step": 6500
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 30.0,
|
| 351 |
+
"eval_accuracy": 0.9285982834369931,
|
| 352 |
+
"eval_loss": 0.4778377413749695,
|
| 353 |
+
"eval_runtime": 5.9957,
|
| 354 |
+
"eval_samples_per_second": 15.177,
|
| 355 |
+
"eval_steps_per_second": 7.672,
|
| 356 |
+
"step": 6600
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 31.0,
|
| 360 |
+
"eval_accuracy": 0.9289527676624451,
|
| 361 |
+
"eval_loss": 0.4756244122982025,
|
| 362 |
+
"eval_runtime": 5.6812,
|
| 363 |
+
"eval_samples_per_second": 16.018,
|
| 364 |
+
"eval_steps_per_second": 8.097,
|
| 365 |
+
"step": 6820
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 31.82,
|
| 369 |
+
"learning_rate": 3.053977272727273e-07,
|
| 370 |
+
"loss": 0.2879,
|
| 371 |
+
"step": 7000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 32.0,
|
| 375 |
+
"eval_accuracy": 0.9289527676624451,
|
| 376 |
+
"eval_loss": 0.47753414511680603,
|
| 377 |
+
"eval_runtime": 5.6774,
|
| 378 |
+
"eval_samples_per_second": 16.029,
|
| 379 |
+
"eval_steps_per_second": 8.102,
|
| 380 |
+
"step": 7040
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 32.0,
|
| 384 |
+
"step": 7040,
|
| 385 |
+
"total_flos": 2.4539422830415053e+17,
|
| 386 |
+
"train_loss": 0.47427067851478405,
|
| 387 |
+
"train_runtime": 10360.152,
|
| 388 |
"train_samples_per_second": 5.442,
|
| 389 |
+
"train_steps_per_second": 0.68
|
| 390 |
}
|
| 391 |
],
|
| 392 |
+
"max_steps": 7040,
|
| 393 |
+
"num_train_epochs": 32,
|
| 394 |
+
"total_flos": 2.4539422830415053e+17,
|
| 395 |
"trial_name": null,
|
| 396 |
"trial_params": null
|
| 397 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3579
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e2dff5363843946656ccbba4034725ba3135ad064c49c180c73190d34e5b0a9d
|
| 3 |
size 3579
|