Pushing deberta-v3-large-hate to hub
Browse files- README.md +115 -0
- added_tokens.json +3 -0
- all_results.json +14 -0
- config.json +43 -0
- eval_results.json +8 -0
- pytorch_model.bin +3 -0
- run_test.sh +1 -0
- run_train.sh +1 -0
- special_tokens_map.json +9 -0
- spm.model +3 -0
- test_results.json +8 -0
- tokenizer.json +0 -0
- tokenizer_config.json +16 -0
- trainer_state.json +865 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,115 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- accuracy
|
| 7 |
+
model-index:
|
| 8 |
+
- name: deberta-v3-large-hate-lr7e-6-gas1-ls0.0
|
| 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 |
+
# deberta-v3-large-hate-lr7e-6-gas1-ls0.0
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [microsoft/deberta-v3-large](https://huggingface.co/microsoft/deberta-v3-large) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.5689
|
| 20 |
+
- Accuracy: 0.8028
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 7e-06
|
| 40 |
+
- train_batch_size: 16
|
| 41 |
+
- eval_batch_size: 16
|
| 42 |
+
- seed: 42
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- lr_scheduler_warmup_steps: 50
|
| 46 |
+
- num_epochs: 10.0
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 51 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 52 |
+
| 0.6362 | 0.18 | 100 | 0.5481 | 0.7197 |
|
| 53 |
+
| 0.4264 | 0.36 | 200 | 0.4550 | 0.8008 |
|
| 54 |
+
| 0.4174 | 0.53 | 300 | 0.4524 | 0.7868 |
|
| 55 |
+
| 0.4197 | 0.71 | 400 | 0.4586 | 0.7918 |
|
| 56 |
+
| 0.3819 | 0.89 | 500 | 0.4368 | 0.8078 |
|
| 57 |
+
| 0.3558 | 1.07 | 600 | 0.4525 | 0.8068 |
|
| 58 |
+
| 0.2982 | 1.24 | 700 | 0.4999 | 0.7928 |
|
| 59 |
+
| 0.2885 | 1.42 | 800 | 0.5129 | 0.8108 |
|
| 60 |
+
| 0.253 | 1.6 | 900 | 0.5873 | 0.8208 |
|
| 61 |
+
| 0.3354 | 1.78 | 1000 | 0.4244 | 0.8178 |
|
| 62 |
+
| 0.3083 | 1.95 | 1100 | 0.4853 | 0.8058 |
|
| 63 |
+
| 0.2301 | 2.13 | 1200 | 0.7209 | 0.8018 |
|
| 64 |
+
| 0.2167 | 2.31 | 1300 | 0.8090 | 0.7778 |
|
| 65 |
+
| 0.1863 | 2.49 | 1400 | 0.6812 | 0.8038 |
|
| 66 |
+
| 0.2181 | 2.66 | 1500 | 0.6958 | 0.8138 |
|
| 67 |
+
| 0.2159 | 2.84 | 1600 | 0.6315 | 0.8118 |
|
| 68 |
+
| 0.1828 | 3.02 | 1700 | 0.7173 | 0.8138 |
|
| 69 |
+
| 0.1287 | 3.2 | 1800 | 0.9081 | 0.8018 |
|
| 70 |
+
| 0.1711 | 3.37 | 1900 | 0.8858 | 0.8068 |
|
| 71 |
+
| 0.1598 | 3.55 | 2000 | 0.7878 | 0.8028 |
|
| 72 |
+
| 0.1467 | 3.73 | 2100 | 0.9003 | 0.7948 |
|
| 73 |
+
| 0.127 | 3.91 | 2200 | 0.9066 | 0.8048 |
|
| 74 |
+
| 0.1134 | 4.09 | 2300 | 0.9646 | 0.8118 |
|
| 75 |
+
| 0.1017 | 4.26 | 2400 | 0.9778 | 0.8048 |
|
| 76 |
+
| 0.085 | 4.44 | 2500 | 1.0529 | 0.8088 |
|
| 77 |
+
| 0.0996 | 4.62 | 2600 | 1.0082 | 0.8058 |
|
| 78 |
+
| 0.1054 | 4.8 | 2700 | 0.9698 | 0.8108 |
|
| 79 |
+
| 0.1375 | 4.97 | 2800 | 0.9334 | 0.8048 |
|
| 80 |
+
| 0.0487 | 5.15 | 2900 | 1.1273 | 0.8108 |
|
| 81 |
+
| 0.0611 | 5.33 | 3000 | 1.1528 | 0.8058 |
|
| 82 |
+
| 0.0668 | 5.51 | 3100 | 1.0148 | 0.8118 |
|
| 83 |
+
| 0.0582 | 5.68 | 3200 | 1.1333 | 0.8108 |
|
| 84 |
+
| 0.0869 | 5.86 | 3300 | 1.0607 | 0.8088 |
|
| 85 |
+
| 0.0623 | 6.04 | 3400 | 1.1880 | 0.8068 |
|
| 86 |
+
| 0.0317 | 6.22 | 3500 | 1.2836 | 0.8008 |
|
| 87 |
+
| 0.0546 | 6.39 | 3600 | 1.2148 | 0.8058 |
|
| 88 |
+
| 0.0486 | 6.57 | 3700 | 1.3348 | 0.8008 |
|
| 89 |
+
| 0.0332 | 6.75 | 3800 | 1.3734 | 0.8018 |
|
| 90 |
+
| 0.051 | 6.93 | 3900 | 1.2966 | 0.7978 |
|
| 91 |
+
| 0.0217 | 7.1 | 4000 | 1.3853 | 0.8048 |
|
| 92 |
+
| 0.0109 | 7.28 | 4100 | 1.4803 | 0.8068 |
|
| 93 |
+
| 0.0345 | 7.46 | 4200 | 1.4906 | 0.7998 |
|
| 94 |
+
| 0.0365 | 7.64 | 4300 | 1.4347 | 0.8028 |
|
| 95 |
+
| 0.0265 | 7.82 | 4400 | 1.3977 | 0.8128 |
|
| 96 |
+
| 0.0257 | 7.99 | 4500 | 1.3705 | 0.8108 |
|
| 97 |
+
| 0.0036 | 8.17 | 4600 | 1.4353 | 0.8168 |
|
| 98 |
+
| 0.0269 | 8.35 | 4700 | 1.4826 | 0.8068 |
|
| 99 |
+
| 0.0231 | 8.53 | 4800 | 1.4811 | 0.8118 |
|
| 100 |
+
| 0.0204 | 8.7 | 4900 | 1.5245 | 0.8028 |
|
| 101 |
+
| 0.0263 | 8.88 | 5000 | 1.5123 | 0.8018 |
|
| 102 |
+
| 0.0138 | 9.06 | 5100 | 1.5113 | 0.8028 |
|
| 103 |
+
| 0.0089 | 9.24 | 5200 | 1.5846 | 0.7978 |
|
| 104 |
+
| 0.029 | 9.41 | 5300 | 1.5362 | 0.8008 |
|
| 105 |
+
| 0.0058 | 9.59 | 5400 | 1.5759 | 0.8018 |
|
| 106 |
+
| 0.0084 | 9.77 | 5500 | 1.5679 | 0.8018 |
|
| 107 |
+
| 0.0065 | 9.95 | 5600 | 1.5683 | 0.8028 |
|
| 108 |
+
|
| 109 |
+
|
| 110 |
+
### Framework versions
|
| 111 |
+
|
| 112 |
+
- Transformers 4.20.0.dev0
|
| 113 |
+
- Pytorch 1.9.0
|
| 114 |
+
- Datasets 2.2.2
|
| 115 |
+
- Tokenizers 0.11.6
|
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[MASK]": 128000
|
| 3 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.8028028011322021,
|
| 4 |
+
"eval_loss": 1.568860650062561,
|
| 5 |
+
"eval_runtime": 6.2145,
|
| 6 |
+
"eval_samples": 999,
|
| 7 |
+
"eval_samples_per_second": 160.753,
|
| 8 |
+
"eval_steps_per_second": 10.138,
|
| 9 |
+
"train_loss": 0.13640729715885533,
|
| 10 |
+
"train_runtime": 2182.3127,
|
| 11 |
+
"train_samples": 9000,
|
| 12 |
+
"train_samples_per_second": 41.241,
|
| 13 |
+
"train_steps_per_second": 2.58
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,43 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "microsoft/deberta-v3-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaV2ForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 1024,
|
| 10 |
+
"id2label": {
|
| 11 |
+
"0": 0,
|
| 12 |
+
"1": 1
|
| 13 |
+
},
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 4096,
|
| 16 |
+
"label2id": {
|
| 17 |
+
"0": 0,
|
| 18 |
+
"1": 1
|
| 19 |
+
},
|
| 20 |
+
"layer_norm_eps": 1e-07,
|
| 21 |
+
"max_position_embeddings": 512,
|
| 22 |
+
"max_relative_positions": -1,
|
| 23 |
+
"model_type": "deberta-v2",
|
| 24 |
+
"norm_rel_ebd": "layer_norm",
|
| 25 |
+
"num_attention_heads": 16,
|
| 26 |
+
"num_hidden_layers": 24,
|
| 27 |
+
"pad_token_id": 0,
|
| 28 |
+
"pooler_dropout": 0,
|
| 29 |
+
"pooler_hidden_act": "gelu",
|
| 30 |
+
"pooler_hidden_size": 1024,
|
| 31 |
+
"pos_att_type": [
|
| 32 |
+
"p2c",
|
| 33 |
+
"c2p"
|
| 34 |
+
],
|
| 35 |
+
"position_biased_input": false,
|
| 36 |
+
"position_buckets": 256,
|
| 37 |
+
"relative_attention": true,
|
| 38 |
+
"share_att_key": true,
|
| 39 |
+
"torch_dtype": "float32",
|
| 40 |
+
"transformers_version": "4.20.0.dev0",
|
| 41 |
+
"type_vocab_size": 0,
|
| 42 |
+
"vocab_size": 128100
|
| 43 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 0.8208208084106445,
|
| 3 |
+
"eval_loss": 0.5872611403465271,
|
| 4 |
+
"eval_runtime": 6.1253,
|
| 5 |
+
"eval_samples": 999,
|
| 6 |
+
"eval_samples_per_second": 163.093,
|
| 7 |
+
"eval_steps_per_second": 10.285
|
| 8 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ab5223d2f36ffa1a5fa517d7faaf2c139d4aa6881beafc0a86c86709ddeb5fbb
|
| 3 |
+
size 1740393387
|
run_test.sh
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
jbsub -queue x86_1h -cores 4+1 -mem 30g -require a100 -o outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0/test.log /dccstor/tslm/envs/anaconda3/envs/tslm-gen/bin/python train_clf.py --model_name_or_path outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0/best_checkpoint --train_file data/tweet_eval/hate/train.csv --validation_file data/tweet_eval/hate/validation.csv --test_file data/tweet_eval/hate/test.csv --do_eval --do_predict --report_to none --per_device_eval_batch_size 16 --max_seq_length 256 --output_dir outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0/best_checkpoint
|
run_train.sh
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
jbsub -queue x86_6h -cores 4+1 -mem 30g -require a100 -o outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0/train.log /dccstor/tslm/envs/anaconda3/envs/tslm-gen/bin/python train_clf.py --model_name_or_path microsoft/deberta-v3-large --train_file data/tweet_eval/hate/train.csv --validation_file data/tweet_eval/hate/validation.csv --do_train --do_eval --per_device_train_batch_size 16 --per_device_eval_batch_size 16 --max_seq_length 256 --learning_rate 7e-6 --output_dir outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0 --evaluation_strategy steps --save_strategy no --warmup_steps 50 --num_train_epochs 10 --overwrite_output_dir --logging_steps 100 --gradient_accumulation_steps 1 --label_smoothing_factor 0.0 --report_to clearml --metric_for_best_model accuracy --logging_dir outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0/tb \; rm -rf outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0/tb \; rm -rf outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0/checkpoint-* \; . outputs/train/tweet_eval2/hate/deberta-v3-large-hate-lr7e-6-gas1-ls0.0/run_test.sh
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"eos_token": "[SEP]",
|
| 5 |
+
"mask_token": "[MASK]",
|
| 6 |
+
"pad_token": "[PAD]",
|
| 7 |
+
"sep_token": "[SEP]",
|
| 8 |
+
"unk_token": "[UNK]"
|
| 9 |
+
}
|
spm.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
|
| 3 |
+
size 2464616
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_accuracy": 0.6127946376800537,
|
| 3 |
+
"eval_loss": 1.4348669052124023,
|
| 4 |
+
"eval_runtime": 17.0067,
|
| 5 |
+
"eval_samples_per_second": 174.637,
|
| 6 |
+
"eval_steps_per_second": 10.937,
|
| 7 |
+
"test_samples": 2970
|
| 8 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_lower_case": false,
|
| 5 |
+
"eos_token": "[SEP]",
|
| 6 |
+
"mask_token": "[MASK]",
|
| 7 |
+
"name_or_path": "microsoft/deberta-v3-large",
|
| 8 |
+
"pad_token": "[PAD]",
|
| 9 |
+
"sep_token": "[SEP]",
|
| 10 |
+
"sp_model_kwargs": {},
|
| 11 |
+
"special_tokens_map_file": null,
|
| 12 |
+
"split_by_punct": false,
|
| 13 |
+
"tokenizer_class": "DebertaV2Tokenizer",
|
| 14 |
+
"unk_token": "[UNK]",
|
| 15 |
+
"vocab_type": "spm"
|
| 16 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,865 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 5630,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.18,
|
| 12 |
+
"learning_rate": 6.937275985663082e-06,
|
| 13 |
+
"loss": 0.6362,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.18,
|
| 18 |
+
"eval_accuracy": 0.7197197079658508,
|
| 19 |
+
"eval_loss": 0.5481122136116028,
|
| 20 |
+
"eval_runtime": 6.2072,
|
| 21 |
+
"eval_samples_per_second": 160.941,
|
| 22 |
+
"eval_steps_per_second": 10.149,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.36,
|
| 27 |
+
"learning_rate": 6.811827956989247e-06,
|
| 28 |
+
"loss": 0.4264,
|
| 29 |
+
"step": 200
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.36,
|
| 33 |
+
"eval_accuracy": 0.8008008003234863,
|
| 34 |
+
"eval_loss": 0.4550396203994751,
|
| 35 |
+
"eval_runtime": 6.2195,
|
| 36 |
+
"eval_samples_per_second": 160.623,
|
| 37 |
+
"eval_steps_per_second": 10.129,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.53,
|
| 42 |
+
"learning_rate": 6.6863799283154114e-06,
|
| 43 |
+
"loss": 0.4174,
|
| 44 |
+
"step": 300
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.53,
|
| 48 |
+
"eval_accuracy": 0.7867867946624756,
|
| 49 |
+
"eval_loss": 0.452409952878952,
|
| 50 |
+
"eval_runtime": 6.2183,
|
| 51 |
+
"eval_samples_per_second": 160.655,
|
| 52 |
+
"eval_steps_per_second": 10.131,
|
| 53 |
+
"step": 300
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.71,
|
| 57 |
+
"learning_rate": 6.560931899641577e-06,
|
| 58 |
+
"loss": 0.4197,
|
| 59 |
+
"step": 400
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.71,
|
| 63 |
+
"eval_accuracy": 0.7917917966842651,
|
| 64 |
+
"eval_loss": 0.4586125910282135,
|
| 65 |
+
"eval_runtime": 6.2441,
|
| 66 |
+
"eval_samples_per_second": 159.991,
|
| 67 |
+
"eval_steps_per_second": 10.09,
|
| 68 |
+
"step": 400
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.89,
|
| 72 |
+
"learning_rate": 6.435483870967742e-06,
|
| 73 |
+
"loss": 0.3819,
|
| 74 |
+
"step": 500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.89,
|
| 78 |
+
"eval_accuracy": 0.8078078031539917,
|
| 79 |
+
"eval_loss": 0.4367608428001404,
|
| 80 |
+
"eval_runtime": 6.2213,
|
| 81 |
+
"eval_samples_per_second": 160.577,
|
| 82 |
+
"eval_steps_per_second": 10.126,
|
| 83 |
+
"step": 500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.07,
|
| 87 |
+
"learning_rate": 6.310035842293907e-06,
|
| 88 |
+
"loss": 0.3558,
|
| 89 |
+
"step": 600
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 1.07,
|
| 93 |
+
"eval_accuracy": 0.8068068027496338,
|
| 94 |
+
"eval_loss": 0.4524727463722229,
|
| 95 |
+
"eval_runtime": 6.2342,
|
| 96 |
+
"eval_samples_per_second": 160.246,
|
| 97 |
+
"eval_steps_per_second": 10.106,
|
| 98 |
+
"step": 600
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.24,
|
| 102 |
+
"learning_rate": 6.184587813620071e-06,
|
| 103 |
+
"loss": 0.2982,
|
| 104 |
+
"step": 700
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.24,
|
| 108 |
+
"eval_accuracy": 0.792792797088623,
|
| 109 |
+
"eval_loss": 0.49992287158966064,
|
| 110 |
+
"eval_runtime": 6.206,
|
| 111 |
+
"eval_samples_per_second": 160.973,
|
| 112 |
+
"eval_steps_per_second": 10.151,
|
| 113 |
+
"step": 700
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.42,
|
| 117 |
+
"learning_rate": 6.059139784946236e-06,
|
| 118 |
+
"loss": 0.2885,
|
| 119 |
+
"step": 800
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.42,
|
| 123 |
+
"eval_accuracy": 0.8108108043670654,
|
| 124 |
+
"eval_loss": 0.5129059553146362,
|
| 125 |
+
"eval_runtime": 6.2199,
|
| 126 |
+
"eval_samples_per_second": 160.613,
|
| 127 |
+
"eval_steps_per_second": 10.129,
|
| 128 |
+
"step": 800
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.6,
|
| 132 |
+
"learning_rate": 5.933691756272401e-06,
|
| 133 |
+
"loss": 0.253,
|
| 134 |
+
"step": 900
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.6,
|
| 138 |
+
"eval_accuracy": 0.8208208084106445,
|
| 139 |
+
"eval_loss": 0.5872611403465271,
|
| 140 |
+
"eval_runtime": 6.2332,
|
| 141 |
+
"eval_samples_per_second": 160.27,
|
| 142 |
+
"eval_steps_per_second": 10.107,
|
| 143 |
+
"step": 900
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.78,
|
| 147 |
+
"learning_rate": 5.8082437275985665e-06,
|
| 148 |
+
"loss": 0.3354,
|
| 149 |
+
"step": 1000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.78,
|
| 153 |
+
"eval_accuracy": 0.8178178071975708,
|
| 154 |
+
"eval_loss": 0.4244420826435089,
|
| 155 |
+
"eval_runtime": 6.2275,
|
| 156 |
+
"eval_samples_per_second": 160.417,
|
| 157 |
+
"eval_steps_per_second": 10.116,
|
| 158 |
+
"step": 1000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.95,
|
| 162 |
+
"learning_rate": 5.682795698924731e-06,
|
| 163 |
+
"loss": 0.3083,
|
| 164 |
+
"step": 1100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.95,
|
| 168 |
+
"eval_accuracy": 0.8058058023452759,
|
| 169 |
+
"eval_loss": 0.4852960705757141,
|
| 170 |
+
"eval_runtime": 6.2193,
|
| 171 |
+
"eval_samples_per_second": 160.63,
|
| 172 |
+
"eval_steps_per_second": 10.13,
|
| 173 |
+
"step": 1100
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 2.13,
|
| 177 |
+
"learning_rate": 5.557347670250896e-06,
|
| 178 |
+
"loss": 0.2301,
|
| 179 |
+
"step": 1200
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.13,
|
| 183 |
+
"eval_accuracy": 0.8018018007278442,
|
| 184 |
+
"eval_loss": 0.7208853960037231,
|
| 185 |
+
"eval_runtime": 6.2021,
|
| 186 |
+
"eval_samples_per_second": 161.075,
|
| 187 |
+
"eval_steps_per_second": 10.158,
|
| 188 |
+
"step": 1200
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.31,
|
| 192 |
+
"learning_rate": 5.431899641577061e-06,
|
| 193 |
+
"loss": 0.2167,
|
| 194 |
+
"step": 1300
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.31,
|
| 198 |
+
"eval_accuracy": 0.7777777910232544,
|
| 199 |
+
"eval_loss": 0.8089737892150879,
|
| 200 |
+
"eval_runtime": 6.2037,
|
| 201 |
+
"eval_samples_per_second": 161.034,
|
| 202 |
+
"eval_steps_per_second": 10.155,
|
| 203 |
+
"step": 1300
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 2.49,
|
| 207 |
+
"learning_rate": 5.306451612903225e-06,
|
| 208 |
+
"loss": 0.1863,
|
| 209 |
+
"step": 1400
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.49,
|
| 213 |
+
"eval_accuracy": 0.8038038015365601,
|
| 214 |
+
"eval_loss": 0.6812323927879333,
|
| 215 |
+
"eval_runtime": 6.2398,
|
| 216 |
+
"eval_samples_per_second": 160.102,
|
| 217 |
+
"eval_steps_per_second": 10.097,
|
| 218 |
+
"step": 1400
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.66,
|
| 222 |
+
"learning_rate": 5.181003584229391e-06,
|
| 223 |
+
"loss": 0.2181,
|
| 224 |
+
"step": 1500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.66,
|
| 228 |
+
"eval_accuracy": 0.8138138055801392,
|
| 229 |
+
"eval_loss": 0.6958026885986328,
|
| 230 |
+
"eval_runtime": 6.2122,
|
| 231 |
+
"eval_samples_per_second": 160.812,
|
| 232 |
+
"eval_steps_per_second": 10.141,
|
| 233 |
+
"step": 1500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.84,
|
| 237 |
+
"learning_rate": 5.0555555555555555e-06,
|
| 238 |
+
"loss": 0.2159,
|
| 239 |
+
"step": 1600
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.84,
|
| 243 |
+
"eval_accuracy": 0.8118118047714233,
|
| 244 |
+
"eval_loss": 0.6314735412597656,
|
| 245 |
+
"eval_runtime": 6.2306,
|
| 246 |
+
"eval_samples_per_second": 160.337,
|
| 247 |
+
"eval_steps_per_second": 10.111,
|
| 248 |
+
"step": 1600
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 3.02,
|
| 252 |
+
"learning_rate": 4.930107526881721e-06,
|
| 253 |
+
"loss": 0.1828,
|
| 254 |
+
"step": 1700
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 3.02,
|
| 258 |
+
"eval_accuracy": 0.8138138055801392,
|
| 259 |
+
"eval_loss": 0.7173236608505249,
|
| 260 |
+
"eval_runtime": 6.2107,
|
| 261 |
+
"eval_samples_per_second": 160.851,
|
| 262 |
+
"eval_steps_per_second": 10.144,
|
| 263 |
+
"step": 1700
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 3.2,
|
| 267 |
+
"learning_rate": 4.804659498207885e-06,
|
| 268 |
+
"loss": 0.1287,
|
| 269 |
+
"step": 1800
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 3.2,
|
| 273 |
+
"eval_accuracy": 0.8018018007278442,
|
| 274 |
+
"eval_loss": 0.9080932140350342,
|
| 275 |
+
"eval_runtime": 6.2027,
|
| 276 |
+
"eval_samples_per_second": 161.06,
|
| 277 |
+
"eval_steps_per_second": 10.157,
|
| 278 |
+
"step": 1800
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.37,
|
| 282 |
+
"learning_rate": 4.67921146953405e-06,
|
| 283 |
+
"loss": 0.1711,
|
| 284 |
+
"step": 1900
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.37,
|
| 288 |
+
"eval_accuracy": 0.8068068027496338,
|
| 289 |
+
"eval_loss": 0.8858422040939331,
|
| 290 |
+
"eval_runtime": 6.2188,
|
| 291 |
+
"eval_samples_per_second": 160.641,
|
| 292 |
+
"eval_steps_per_second": 10.131,
|
| 293 |
+
"step": 1900
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.55,
|
| 297 |
+
"learning_rate": 4.553763440860215e-06,
|
| 298 |
+
"loss": 0.1598,
|
| 299 |
+
"step": 2000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.55,
|
| 303 |
+
"eval_accuracy": 0.8028028011322021,
|
| 304 |
+
"eval_loss": 0.7877860069274902,
|
| 305 |
+
"eval_runtime": 6.2062,
|
| 306 |
+
"eval_samples_per_second": 160.967,
|
| 307 |
+
"eval_steps_per_second": 10.151,
|
| 308 |
+
"step": 2000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 3.73,
|
| 312 |
+
"learning_rate": 4.42831541218638e-06,
|
| 313 |
+
"loss": 0.1467,
|
| 314 |
+
"step": 2100
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.73,
|
| 318 |
+
"eval_accuracy": 0.7947947978973389,
|
| 319 |
+
"eval_loss": 0.900332510471344,
|
| 320 |
+
"eval_runtime": 6.2358,
|
| 321 |
+
"eval_samples_per_second": 160.203,
|
| 322 |
+
"eval_steps_per_second": 10.103,
|
| 323 |
+
"step": 2100
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.91,
|
| 327 |
+
"learning_rate": 4.302867383512545e-06,
|
| 328 |
+
"loss": 0.127,
|
| 329 |
+
"step": 2200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.91,
|
| 333 |
+
"eval_accuracy": 0.804804801940918,
|
| 334 |
+
"eval_loss": 0.9066368341445923,
|
| 335 |
+
"eval_runtime": 6.2129,
|
| 336 |
+
"eval_samples_per_second": 160.795,
|
| 337 |
+
"eval_steps_per_second": 10.14,
|
| 338 |
+
"step": 2200
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 4.09,
|
| 342 |
+
"learning_rate": 4.17741935483871e-06,
|
| 343 |
+
"loss": 0.1134,
|
| 344 |
+
"step": 2300
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 4.09,
|
| 348 |
+
"eval_accuracy": 0.8118118047714233,
|
| 349 |
+
"eval_loss": 0.9645766615867615,
|
| 350 |
+
"eval_runtime": 6.2157,
|
| 351 |
+
"eval_samples_per_second": 160.721,
|
| 352 |
+
"eval_steps_per_second": 10.136,
|
| 353 |
+
"step": 2300
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 4.26,
|
| 357 |
+
"learning_rate": 4.051971326164874e-06,
|
| 358 |
+
"loss": 0.1017,
|
| 359 |
+
"step": 2400
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 4.26,
|
| 363 |
+
"eval_accuracy": 0.804804801940918,
|
| 364 |
+
"eval_loss": 0.9778422713279724,
|
| 365 |
+
"eval_runtime": 6.2303,
|
| 366 |
+
"eval_samples_per_second": 160.346,
|
| 367 |
+
"eval_steps_per_second": 10.112,
|
| 368 |
+
"step": 2400
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 4.44,
|
| 372 |
+
"learning_rate": 3.926523297491039e-06,
|
| 373 |
+
"loss": 0.085,
|
| 374 |
+
"step": 2500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 4.44,
|
| 378 |
+
"eval_accuracy": 0.8088088035583496,
|
| 379 |
+
"eval_loss": 1.0528582334518433,
|
| 380 |
+
"eval_runtime": 6.238,
|
| 381 |
+
"eval_samples_per_second": 160.149,
|
| 382 |
+
"eval_steps_per_second": 10.099,
|
| 383 |
+
"step": 2500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.62,
|
| 387 |
+
"learning_rate": 3.801075268817204e-06,
|
| 388 |
+
"loss": 0.0996,
|
| 389 |
+
"step": 2600
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 4.62,
|
| 393 |
+
"eval_accuracy": 0.8058058023452759,
|
| 394 |
+
"eval_loss": 1.0082268714904785,
|
| 395 |
+
"eval_runtime": 6.2065,
|
| 396 |
+
"eval_samples_per_second": 160.961,
|
| 397 |
+
"eval_steps_per_second": 10.151,
|
| 398 |
+
"step": 2600
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.8,
|
| 402 |
+
"learning_rate": 3.6756272401433694e-06,
|
| 403 |
+
"loss": 0.1054,
|
| 404 |
+
"step": 2700
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 4.8,
|
| 408 |
+
"eval_accuracy": 0.8108108043670654,
|
| 409 |
+
"eval_loss": 0.9697705507278442,
|
| 410 |
+
"eval_runtime": 6.2348,
|
| 411 |
+
"eval_samples_per_second": 160.231,
|
| 412 |
+
"eval_steps_per_second": 10.105,
|
| 413 |
+
"step": 2700
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.97,
|
| 417 |
+
"learning_rate": 3.5501792114695336e-06,
|
| 418 |
+
"loss": 0.1375,
|
| 419 |
+
"step": 2800
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 4.97,
|
| 423 |
+
"eval_accuracy": 0.804804801940918,
|
| 424 |
+
"eval_loss": 0.9333746433258057,
|
| 425 |
+
"eval_runtime": 6.2109,
|
| 426 |
+
"eval_samples_per_second": 160.846,
|
| 427 |
+
"eval_steps_per_second": 10.143,
|
| 428 |
+
"step": 2800
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 5.15,
|
| 432 |
+
"learning_rate": 3.4247311827956988e-06,
|
| 433 |
+
"loss": 0.0487,
|
| 434 |
+
"step": 2900
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 5.15,
|
| 438 |
+
"eval_accuracy": 0.8108108043670654,
|
| 439 |
+
"eval_loss": 1.1273365020751953,
|
| 440 |
+
"eval_runtime": 6.2065,
|
| 441 |
+
"eval_samples_per_second": 160.961,
|
| 442 |
+
"eval_steps_per_second": 10.151,
|
| 443 |
+
"step": 2900
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 5.33,
|
| 447 |
+
"learning_rate": 3.299283154121864e-06,
|
| 448 |
+
"loss": 0.0611,
|
| 449 |
+
"step": 3000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 5.33,
|
| 453 |
+
"eval_accuracy": 0.8058058023452759,
|
| 454 |
+
"eval_loss": 1.1528337001800537,
|
| 455 |
+
"eval_runtime": 6.2119,
|
| 456 |
+
"eval_samples_per_second": 160.821,
|
| 457 |
+
"eval_steps_per_second": 10.142,
|
| 458 |
+
"step": 3000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 5.51,
|
| 462 |
+
"learning_rate": 3.1738351254480286e-06,
|
| 463 |
+
"loss": 0.0668,
|
| 464 |
+
"step": 3100
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 5.51,
|
| 468 |
+
"eval_accuracy": 0.8118118047714233,
|
| 469 |
+
"eval_loss": 1.0147671699523926,
|
| 470 |
+
"eval_runtime": 6.2218,
|
| 471 |
+
"eval_samples_per_second": 160.564,
|
| 472 |
+
"eval_steps_per_second": 10.126,
|
| 473 |
+
"step": 3100
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 5.68,
|
| 477 |
+
"learning_rate": 3.0483870967741937e-06,
|
| 478 |
+
"loss": 0.0582,
|
| 479 |
+
"step": 3200
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 5.68,
|
| 483 |
+
"eval_accuracy": 0.8108108043670654,
|
| 484 |
+
"eval_loss": 1.1332666873931885,
|
| 485 |
+
"eval_runtime": 6.2186,
|
| 486 |
+
"eval_samples_per_second": 160.648,
|
| 487 |
+
"eval_steps_per_second": 10.131,
|
| 488 |
+
"step": 3200
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 5.86,
|
| 492 |
+
"learning_rate": 2.9229390681003584e-06,
|
| 493 |
+
"loss": 0.0869,
|
| 494 |
+
"step": 3300
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 5.86,
|
| 498 |
+
"eval_accuracy": 0.8088088035583496,
|
| 499 |
+
"eval_loss": 1.060727596282959,
|
| 500 |
+
"eval_runtime": 6.1932,
|
| 501 |
+
"eval_samples_per_second": 161.305,
|
| 502 |
+
"eval_steps_per_second": 10.172,
|
| 503 |
+
"step": 3300
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 6.04,
|
| 507 |
+
"learning_rate": 2.797491039426523e-06,
|
| 508 |
+
"loss": 0.0623,
|
| 509 |
+
"step": 3400
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 6.04,
|
| 513 |
+
"eval_accuracy": 0.8068068027496338,
|
| 514 |
+
"eval_loss": 1.1880476474761963,
|
| 515 |
+
"eval_runtime": 6.2192,
|
| 516 |
+
"eval_samples_per_second": 160.631,
|
| 517 |
+
"eval_steps_per_second": 10.13,
|
| 518 |
+
"step": 3400
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 6.22,
|
| 522 |
+
"learning_rate": 2.6720430107526883e-06,
|
| 523 |
+
"loss": 0.0317,
|
| 524 |
+
"step": 3500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 6.22,
|
| 528 |
+
"eval_accuracy": 0.8008008003234863,
|
| 529 |
+
"eval_loss": 1.2836244106292725,
|
| 530 |
+
"eval_runtime": 6.2079,
|
| 531 |
+
"eval_samples_per_second": 160.925,
|
| 532 |
+
"eval_steps_per_second": 10.148,
|
| 533 |
+
"step": 3500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 6.39,
|
| 537 |
+
"learning_rate": 2.546594982078853e-06,
|
| 538 |
+
"loss": 0.0546,
|
| 539 |
+
"step": 3600
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 6.39,
|
| 543 |
+
"eval_accuracy": 0.8058058023452759,
|
| 544 |
+
"eval_loss": 1.2147704362869263,
|
| 545 |
+
"eval_runtime": 6.2243,
|
| 546 |
+
"eval_samples_per_second": 160.501,
|
| 547 |
+
"eval_steps_per_second": 10.122,
|
| 548 |
+
"step": 3600
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 6.57,
|
| 552 |
+
"learning_rate": 2.4211469534050177e-06,
|
| 553 |
+
"loss": 0.0486,
|
| 554 |
+
"step": 3700
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 6.57,
|
| 558 |
+
"eval_accuracy": 0.8008008003234863,
|
| 559 |
+
"eval_loss": 1.334807276725769,
|
| 560 |
+
"eval_runtime": 6.1963,
|
| 561 |
+
"eval_samples_per_second": 161.225,
|
| 562 |
+
"eval_steps_per_second": 10.167,
|
| 563 |
+
"step": 3700
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 6.75,
|
| 567 |
+
"learning_rate": 2.2956989247311828e-06,
|
| 568 |
+
"loss": 0.0332,
|
| 569 |
+
"step": 3800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 6.75,
|
| 573 |
+
"eval_accuracy": 0.8018018007278442,
|
| 574 |
+
"eval_loss": 1.3734461069107056,
|
| 575 |
+
"eval_runtime": 6.3321,
|
| 576 |
+
"eval_samples_per_second": 157.768,
|
| 577 |
+
"eval_steps_per_second": 9.949,
|
| 578 |
+
"step": 3800
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 6.93,
|
| 582 |
+
"learning_rate": 2.1702508960573475e-06,
|
| 583 |
+
"loss": 0.051,
|
| 584 |
+
"step": 3900
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 6.93,
|
| 588 |
+
"eval_accuracy": 0.7977977991104126,
|
| 589 |
+
"eval_loss": 1.2966439723968506,
|
| 590 |
+
"eval_runtime": 6.2073,
|
| 591 |
+
"eval_samples_per_second": 160.94,
|
| 592 |
+
"eval_steps_per_second": 10.149,
|
| 593 |
+
"step": 3900
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 7.1,
|
| 597 |
+
"learning_rate": 2.044802867383512e-06,
|
| 598 |
+
"loss": 0.0217,
|
| 599 |
+
"step": 4000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 7.1,
|
| 603 |
+
"eval_accuracy": 0.804804801940918,
|
| 604 |
+
"eval_loss": 1.385273814201355,
|
| 605 |
+
"eval_runtime": 6.2117,
|
| 606 |
+
"eval_samples_per_second": 160.826,
|
| 607 |
+
"eval_steps_per_second": 10.142,
|
| 608 |
+
"step": 4000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 7.28,
|
| 612 |
+
"learning_rate": 1.9193548387096773e-06,
|
| 613 |
+
"loss": 0.0109,
|
| 614 |
+
"step": 4100
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 7.28,
|
| 618 |
+
"eval_accuracy": 0.8068068027496338,
|
| 619 |
+
"eval_loss": 1.480326533317566,
|
| 620 |
+
"eval_runtime": 6.2106,
|
| 621 |
+
"eval_samples_per_second": 160.854,
|
| 622 |
+
"eval_steps_per_second": 10.144,
|
| 623 |
+
"step": 4100
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 7.46,
|
| 627 |
+
"learning_rate": 1.793906810035842e-06,
|
| 628 |
+
"loss": 0.0345,
|
| 629 |
+
"step": 4200
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 7.46,
|
| 633 |
+
"eval_accuracy": 0.7997997999191284,
|
| 634 |
+
"eval_loss": 1.4906260967254639,
|
| 635 |
+
"eval_runtime": 6.2002,
|
| 636 |
+
"eval_samples_per_second": 161.124,
|
| 637 |
+
"eval_steps_per_second": 10.161,
|
| 638 |
+
"step": 4200
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 7.64,
|
| 642 |
+
"learning_rate": 1.6684587813620071e-06,
|
| 643 |
+
"loss": 0.0365,
|
| 644 |
+
"step": 4300
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 7.64,
|
| 648 |
+
"eval_accuracy": 0.8028028011322021,
|
| 649 |
+
"eval_loss": 1.4347106218338013,
|
| 650 |
+
"eval_runtime": 6.2133,
|
| 651 |
+
"eval_samples_per_second": 160.783,
|
| 652 |
+
"eval_steps_per_second": 10.139,
|
| 653 |
+
"step": 4300
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 7.82,
|
| 657 |
+
"learning_rate": 1.543010752688172e-06,
|
| 658 |
+
"loss": 0.0265,
|
| 659 |
+
"step": 4400
|
| 660 |
+
},
|
| 661 |
+
{
|
| 662 |
+
"epoch": 7.82,
|
| 663 |
+
"eval_accuracy": 0.8128128051757812,
|
| 664 |
+
"eval_loss": 1.3976863622665405,
|
| 665 |
+
"eval_runtime": 6.224,
|
| 666 |
+
"eval_samples_per_second": 160.508,
|
| 667 |
+
"eval_steps_per_second": 10.122,
|
| 668 |
+
"step": 4400
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 7.99,
|
| 672 |
+
"learning_rate": 1.417562724014337e-06,
|
| 673 |
+
"loss": 0.0257,
|
| 674 |
+
"step": 4500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 7.99,
|
| 678 |
+
"eval_accuracy": 0.8108108043670654,
|
| 679 |
+
"eval_loss": 1.370467185974121,
|
| 680 |
+
"eval_runtime": 6.2313,
|
| 681 |
+
"eval_samples_per_second": 160.321,
|
| 682 |
+
"eval_steps_per_second": 10.11,
|
| 683 |
+
"step": 4500
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 8.17,
|
| 687 |
+
"learning_rate": 1.2921146953405017e-06,
|
| 688 |
+
"loss": 0.0036,
|
| 689 |
+
"step": 4600
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 8.17,
|
| 693 |
+
"eval_accuracy": 0.8168168067932129,
|
| 694 |
+
"eval_loss": 1.4352822303771973,
|
| 695 |
+
"eval_runtime": 6.2072,
|
| 696 |
+
"eval_samples_per_second": 160.943,
|
| 697 |
+
"eval_steps_per_second": 10.15,
|
| 698 |
+
"step": 4600
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 8.35,
|
| 702 |
+
"learning_rate": 1.1666666666666666e-06,
|
| 703 |
+
"loss": 0.0269,
|
| 704 |
+
"step": 4700
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 8.35,
|
| 708 |
+
"eval_accuracy": 0.8068068027496338,
|
| 709 |
+
"eval_loss": 1.4826140403747559,
|
| 710 |
+
"eval_runtime": 6.2178,
|
| 711 |
+
"eval_samples_per_second": 160.669,
|
| 712 |
+
"eval_steps_per_second": 10.132,
|
| 713 |
+
"step": 4700
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 8.53,
|
| 717 |
+
"learning_rate": 1.0412186379928315e-06,
|
| 718 |
+
"loss": 0.0231,
|
| 719 |
+
"step": 4800
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 8.53,
|
| 723 |
+
"eval_accuracy": 0.8118118047714233,
|
| 724 |
+
"eval_loss": 1.4810999631881714,
|
| 725 |
+
"eval_runtime": 6.3061,
|
| 726 |
+
"eval_samples_per_second": 158.417,
|
| 727 |
+
"eval_steps_per_second": 9.99,
|
| 728 |
+
"step": 4800
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 8.7,
|
| 732 |
+
"learning_rate": 9.157706093189965e-07,
|
| 733 |
+
"loss": 0.0204,
|
| 734 |
+
"step": 4900
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 8.7,
|
| 738 |
+
"eval_accuracy": 0.8028028011322021,
|
| 739 |
+
"eval_loss": 1.5245323181152344,
|
| 740 |
+
"eval_runtime": 6.2057,
|
| 741 |
+
"eval_samples_per_second": 160.982,
|
| 742 |
+
"eval_steps_per_second": 10.152,
|
| 743 |
+
"step": 4900
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 8.88,
|
| 747 |
+
"learning_rate": 7.903225806451612e-07,
|
| 748 |
+
"loss": 0.0263,
|
| 749 |
+
"step": 5000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 8.88,
|
| 753 |
+
"eval_accuracy": 0.8018018007278442,
|
| 754 |
+
"eval_loss": 1.5123308897018433,
|
| 755 |
+
"eval_runtime": 6.2053,
|
| 756 |
+
"eval_samples_per_second": 160.991,
|
| 757 |
+
"eval_steps_per_second": 10.153,
|
| 758 |
+
"step": 5000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 9.06,
|
| 762 |
+
"learning_rate": 6.648745519713261e-07,
|
| 763 |
+
"loss": 0.0138,
|
| 764 |
+
"step": 5100
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 9.06,
|
| 768 |
+
"eval_accuracy": 0.8028028011322021,
|
| 769 |
+
"eval_loss": 1.51128089427948,
|
| 770 |
+
"eval_runtime": 6.2898,
|
| 771 |
+
"eval_samples_per_second": 158.83,
|
| 772 |
+
"eval_steps_per_second": 10.016,
|
| 773 |
+
"step": 5100
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 9.24,
|
| 777 |
+
"learning_rate": 5.39426523297491e-07,
|
| 778 |
+
"loss": 0.0089,
|
| 779 |
+
"step": 5200
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 9.24,
|
| 783 |
+
"eval_accuracy": 0.7977977991104126,
|
| 784 |
+
"eval_loss": 1.5846397876739502,
|
| 785 |
+
"eval_runtime": 6.2124,
|
| 786 |
+
"eval_samples_per_second": 160.808,
|
| 787 |
+
"eval_steps_per_second": 10.141,
|
| 788 |
+
"step": 5200
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 9.41,
|
| 792 |
+
"learning_rate": 4.1397849462365595e-07,
|
| 793 |
+
"loss": 0.029,
|
| 794 |
+
"step": 5300
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 9.41,
|
| 798 |
+
"eval_accuracy": 0.8008008003234863,
|
| 799 |
+
"eval_loss": 1.5361814498901367,
|
| 800 |
+
"eval_runtime": 6.2541,
|
| 801 |
+
"eval_samples_per_second": 159.736,
|
| 802 |
+
"eval_steps_per_second": 10.073,
|
| 803 |
+
"step": 5300
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 9.59,
|
| 807 |
+
"learning_rate": 2.8853046594982076e-07,
|
| 808 |
+
"loss": 0.0058,
|
| 809 |
+
"step": 5400
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 9.59,
|
| 813 |
+
"eval_accuracy": 0.8018018007278442,
|
| 814 |
+
"eval_loss": 1.5759379863739014,
|
| 815 |
+
"eval_runtime": 6.221,
|
| 816 |
+
"eval_samples_per_second": 160.585,
|
| 817 |
+
"eval_steps_per_second": 10.127,
|
| 818 |
+
"step": 5400
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 9.77,
|
| 822 |
+
"learning_rate": 1.6308243727598568e-07,
|
| 823 |
+
"loss": 0.0084,
|
| 824 |
+
"step": 5500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 9.77,
|
| 828 |
+
"eval_accuracy": 0.8018018007278442,
|
| 829 |
+
"eval_loss": 1.5678976774215698,
|
| 830 |
+
"eval_runtime": 6.2009,
|
| 831 |
+
"eval_samples_per_second": 161.105,
|
| 832 |
+
"eval_steps_per_second": 10.16,
|
| 833 |
+
"step": 5500
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 9.95,
|
| 837 |
+
"learning_rate": 3.763440860215054e-08,
|
| 838 |
+
"loss": 0.0065,
|
| 839 |
+
"step": 5600
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 9.95,
|
| 843 |
+
"eval_accuracy": 0.8028028011322021,
|
| 844 |
+
"eval_loss": 1.568334937095642,
|
| 845 |
+
"eval_runtime": 6.2439,
|
| 846 |
+
"eval_samples_per_second": 159.996,
|
| 847 |
+
"eval_steps_per_second": 10.09,
|
| 848 |
+
"step": 5600
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 10.0,
|
| 852 |
+
"step": 5630,
|
| 853 |
+
"total_flos": 4.193719446528e+16,
|
| 854 |
+
"train_loss": 0.13640729715885533,
|
| 855 |
+
"train_runtime": 2182.3127,
|
| 856 |
+
"train_samples_per_second": 41.241,
|
| 857 |
+
"train_steps_per_second": 2.58
|
| 858 |
+
}
|
| 859 |
+
],
|
| 860 |
+
"max_steps": 5630,
|
| 861 |
+
"num_train_epochs": 10,
|
| 862 |
+
"total_flos": 4.193719446528e+16,
|
| 863 |
+
"trial_name": null,
|
| 864 |
+
"trial_params": null
|
| 865 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:905229a66e041e28383381e29a8c40d438d511982854d7195ae501648f5d83ad
|
| 3 |
+
size 3311
|