Commit ·
3cbed44
1
Parent(s): 2f95de8
Model update
Browse files- README.md +16 -17
- all_results.json +13 -13
- config.json +1 -1
- eval_results.json +8 -8
- generation_config.json +1 -1
- pytorch_model-00001-of-00002.bin +1 -1
- pytorch_model-00002-of-00002.bin +1 -1
- runs/{Mar26_04-01-41_panacea/1679817763.2045138/events.out.tfevents.1679817763.panacea.358302.1 → May20_16-39-41_panacea/1684615241.5386765/events.out.tfevents.1684615241.panacea.344900.1} +2 -2
- runs/{Mar26_04-01-41_panacea/events.out.tfevents.1679817763.panacea.358302.0 → May20_16-39-41_panacea/events.out.tfevents.1684615241.panacea.344900.0} +2 -2
- runs/{Mar26_04-01-41_panacea/events.out.tfevents.1679871090.panacea.358302.2 → May20_16-39-41_panacea/events.out.tfevents.1684662404.panacea.344900.2} +1 -1
- tokenizer_config.json +1 -1
- train_results.json +6 -6
- trainer_state.json +693 -564
- 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 [facebook/opt-2.7b](https://huggingface.co/facebook/opt-2.7b) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 2.
|
| 20 |
-
- Accuracy: 0.
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
@@ -36,33 +36,32 @@ 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
|
| 43 |
-
- gradient_accumulation_steps:
|
| 44 |
-
- total_train_batch_size:
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
- lr_scheduler_warmup_ratio: 0.1
|
| 48 |
-
- num_epochs:
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 54 |
-
| 2.
|
| 55 |
-
| 2.
|
| 56 |
-
| 1.
|
| 57 |
-
| 1.
|
| 58 |
-
| 1.
|
| 59 |
-
| 1.
|
| 60 |
-
| 1.3551 | 3.51 | 1141 | 2.0624 | 0.7373 |
|
| 61 |
|
| 62 |
|
| 63 |
### Framework versions
|
| 64 |
|
| 65 |
-
- Transformers 4.
|
| 66 |
-
- Pytorch
|
| 67 |
- Datasets 2.7.1
|
| 68 |
- Tokenizers 0.12.1
|
|
|
|
| 5 |
metrics:
|
| 6 |
- accuracy
|
| 7 |
model-index:
|
| 8 |
+
- name: opt-2.7b-realtime-chat-v2
|
| 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 |
+
# opt-2.7b-realtime-chat-v2
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [facebook/opt-2.7b](https://huggingface.co/facebook/opt-2.7b) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 2.0888
|
| 20 |
+
- Accuracy: 0.6870
|
| 21 |
|
| 22 |
## Model description
|
| 23 |
|
|
|
|
| 36 |
### Training hyperparameters
|
| 37 |
|
| 38 |
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 3e-05
|
| 40 |
- train_batch_size: 1
|
| 41 |
- eval_batch_size: 1
|
| 42 |
- seed: 42
|
| 43 |
+
- gradient_accumulation_steps: 128
|
| 44 |
+
- total_train_batch_size: 128
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
- lr_scheduler_warmup_ratio: 0.1
|
| 48 |
+
- num_epochs: 3.0
|
| 49 |
|
| 50 |
### Training results
|
| 51 |
|
| 52 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 53 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 54 |
+
| 2.0974 | 0.5 | 51 | 2.1267 | 0.6826 |
|
| 55 |
+
| 2.0842 | 1.0 | 102 | 2.0968 | 0.6859 |
|
| 56 |
+
| 1.9624 | 1.49 | 153 | 2.0936 | 0.6863 |
|
| 57 |
+
| 1.9476 | 1.99 | 204 | 2.0888 | 0.6870 |
|
| 58 |
+
| 1.888 | 2.49 | 255 | 2.0993 | 0.6864 |
|
| 59 |
+
| 1.8687 | 2.99 | 306 | 2.0994 | 0.6865 |
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
| 63 |
|
| 64 |
+
- Transformers 4.28.1
|
| 65 |
+
- Pytorch 2.0.1+cu118
|
| 66 |
- Datasets 2.7.1
|
| 67 |
- Tokenizers 0.12.1
|
all_results.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 2.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples":
|
| 7 |
-
"eval_samples_per_second": 3.
|
| 8 |
-
"eval_steps_per_second": 3.
|
| 9 |
-
"perplexity":
|
| 10 |
-
"train_loss":
|
| 11 |
-
"train_runtime":
|
| 12 |
-
"train_samples":
|
| 13 |
-
"train_samples_per_second": 0.
|
| 14 |
-
"train_steps_per_second": 0.
|
| 15 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.99,
|
| 3 |
+
"eval_accuracy": 0.6870071488415761,
|
| 4 |
+
"eval_loss": 2.0888001918792725,
|
| 5 |
+
"eval_runtime": 509.5975,
|
| 6 |
+
"eval_samples": 1796,
|
| 7 |
+
"eval_samples_per_second": 3.524,
|
| 8 |
+
"eval_steps_per_second": 3.524,
|
| 9 |
+
"perplexity": 8.075220634386367,
|
| 10 |
+
"train_loss": 2.010221361334807,
|
| 11 |
+
"train_runtime": 46544.0603,
|
| 12 |
+
"train_samples": 13111,
|
| 13 |
+
"train_samples_per_second": 0.845,
|
| 14 |
+
"train_steps_per_second": 0.007
|
| 15 |
}
|
config.json
CHANGED
|
@@ -24,7 +24,7 @@
|
|
| 24 |
"pad_token_id": 1,
|
| 25 |
"prefix": "</s>",
|
| 26 |
"torch_dtype": "float32",
|
| 27 |
-
"transformers_version": "4.
|
| 28 |
"use_cache": true,
|
| 29 |
"vocab_size": 50265,
|
| 30 |
"word_embed_proj_dim": 2560
|
|
|
|
| 24 |
"pad_token_id": 1,
|
| 25 |
"prefix": "</s>",
|
| 26 |
"torch_dtype": "float32",
|
| 27 |
+
"transformers_version": "4.28.1",
|
| 28 |
"use_cache": true,
|
| 29 |
"vocab_size": 50265,
|
| 30 |
"word_embed_proj_dim": 2560
|
eval_results.json
CHANGED
|
@@ -1,10 +1,10 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 2.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples":
|
| 7 |
-
"eval_samples_per_second": 3.
|
| 8 |
-
"eval_steps_per_second": 3.
|
| 9 |
-
"perplexity":
|
| 10 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.99,
|
| 3 |
+
"eval_accuracy": 0.6870071488415761,
|
| 4 |
+
"eval_loss": 2.0888001918792725,
|
| 5 |
+
"eval_runtime": 509.5975,
|
| 6 |
+
"eval_samples": 1796,
|
| 7 |
+
"eval_samples_per_second": 3.524,
|
| 8 |
+
"eval_steps_per_second": 3.524,
|
| 9 |
+
"perplexity": 8.075220634386367
|
| 10 |
}
|
generation_config.json
CHANGED
|
@@ -3,5 +3,5 @@
|
|
| 3 |
"bos_token_id": 2,
|
| 4 |
"eos_token_id": 2,
|
| 5 |
"pad_token_id": 1,
|
| 6 |
-
"transformers_version": "4.
|
| 7 |
}
|
|
|
|
| 3 |
"bos_token_id": 2,
|
| 4 |
"eos_token_id": 2,
|
| 5 |
"pad_token_id": 1,
|
| 6 |
+
"transformers_version": "4.28.1"
|
| 7 |
}
|
pytorch_model-00001-of-00002.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 9977078723
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5663e8b1081fbaf6440fd6ab38f293ffa5eb67c0af62a2e7775192557c0eefed
|
| 3 |
size 9977078723
|
pytorch_model-00002-of-00002.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1144136929
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e483afa16baf460b6d548b4fcefe15a95b96faca2a1d00300508694bf9535f90
|
| 3 |
size 1144136929
|
runs/{Mar26_04-01-41_panacea/1679817763.2045138/events.out.tfevents.1679817763.panacea.358302.1 → May20_16-39-41_panacea/1684615241.5386765/events.out.tfevents.1684615241.panacea.344900.1}
RENAMED
|
@@ -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 sha256:72486d16f01267fe4033873a7d5e8848b301f38f611bbcf77b5a2ffc8b55e3ae
|
| 3 |
+
size 5859
|
runs/{Mar26_04-01-41_panacea/events.out.tfevents.1679817763.panacea.358302.0 → May20_16-39-41_panacea/events.out.tfevents.1684615241.panacea.344900.0}
RENAMED
|
@@ -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:eba7fbc4b3b305a059bd4a5914673a837e6936dc8d13dcb4a52646a703572f66
|
| 3 |
+
size 30178
|
runs/{Mar26_04-01-41_panacea/events.out.tfevents.1679871090.panacea.358302.2 → May20_16-39-41_panacea/events.out.tfevents.1684662404.panacea.344900.2}
RENAMED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 363
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0dfc715017543caa195a56d8608c6c94bc4548f9b212793816cb9569e4dd3179
|
| 3 |
size 363
|
tokenizer_config.json
CHANGED
|
@@ -9,6 +9,7 @@
|
|
| 9 |
"rstrip": false,
|
| 10 |
"single_word": false
|
| 11 |
},
|
|
|
|
| 12 |
"eos_token": {
|
| 13 |
"__type": "AddedToken",
|
| 14 |
"content": "</s>",
|
|
@@ -27,7 +28,6 @@
|
|
| 27 |
"rstrip": false,
|
| 28 |
"single_word": false
|
| 29 |
},
|
| 30 |
-
"special_tokens_map_file": null,
|
| 31 |
"tokenizer_class": "GPT2Tokenizer",
|
| 32 |
"unk_token": {
|
| 33 |
"__type": "AddedToken",
|
|
|
|
| 9 |
"rstrip": false,
|
| 10 |
"single_word": false
|
| 11 |
},
|
| 12 |
+
"clean_up_tokenization_spaces": true,
|
| 13 |
"eos_token": {
|
| 14 |
"__type": "AddedToken",
|
| 15 |
"content": "</s>",
|
|
|
|
| 28 |
"rstrip": false,
|
| 29 |
"single_word": false
|
| 30 |
},
|
|
|
|
| 31 |
"tokenizer_class": "GPT2Tokenizer",
|
| 32 |
"unk_token": {
|
| 33 |
"__type": "AddedToken",
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
-
"train_samples":
|
| 6 |
-
"train_samples_per_second": 0.
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 2.99,
|
| 3 |
+
"train_loss": 2.010221361334807,
|
| 4 |
+
"train_runtime": 46544.0603,
|
| 5 |
+
"train_samples": 13111,
|
| 6 |
+
"train_samples_per_second": 0.845,
|
| 7 |
+
"train_steps_per_second": 0.007
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,868 +1,997 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric":
|
| 3 |
-
"best_model_checkpoint": "
|
| 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": 0.
|
| 12 |
-
"learning_rate":
|
| 13 |
-
"loss": 2.
|
| 14 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
},
|
| 16 |
{
|
| 17 |
"epoch": 0.06,
|
| 18 |
-
"learning_rate":
|
| 19 |
-
"loss": 2.
|
| 20 |
-
"step":
|
| 21 |
},
|
| 22 |
{
|
| 23 |
-
"epoch": 0.
|
| 24 |
-
"learning_rate":
|
| 25 |
-
"loss": 2.
|
| 26 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
},
|
| 28 |
{
|
| 29 |
"epoch": 0.12,
|
| 30 |
-
"learning_rate": 1.
|
| 31 |
-
"loss": 2.
|
| 32 |
-
"step":
|
| 33 |
},
|
| 34 |
{
|
| 35 |
-
"epoch": 0.
|
| 36 |
-
"learning_rate": 1.
|
| 37 |
-
"loss": 2.
|
| 38 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.18,
|
| 42 |
-
"learning_rate":
|
| 43 |
-
"loss": 2.
|
| 44 |
-
"step":
|
| 45 |
},
|
| 46 |
{
|
| 47 |
-
"epoch": 0.
|
| 48 |
-
"learning_rate":
|
| 49 |
-
"loss": 2.
|
| 50 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 51 |
},
|
| 52 |
{
|
| 53 |
"epoch": 0.25,
|
| 54 |
-
"learning_rate":
|
| 55 |
-
"loss": 2.
|
| 56 |
-
"step":
|
| 57 |
},
|
| 58 |
{
|
| 59 |
-
"epoch": 0.
|
| 60 |
-
"learning_rate":
|
| 61 |
-
"loss": 2.
|
| 62 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 63 |
},
|
| 64 |
{
|
| 65 |
"epoch": 0.31,
|
| 66 |
-
"learning_rate":
|
| 67 |
-
"loss": 2.
|
| 68 |
-
"step":
|
| 69 |
},
|
| 70 |
{
|
| 71 |
-
"epoch": 0.
|
| 72 |
-
"learning_rate":
|
| 73 |
-
"loss": 2.
|
| 74 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"epoch": 0.37,
|
| 78 |
-
"learning_rate":
|
| 79 |
-
"loss": 2.
|
| 80 |
-
"step":
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"epoch": 0.
|
| 84 |
-
"learning_rate":
|
| 85 |
-
"loss": 2.
|
| 86 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 87 |
},
|
| 88 |
{
|
| 89 |
"epoch": 0.43,
|
| 90 |
-
"learning_rate":
|
| 91 |
-
"loss": 2.
|
| 92 |
-
"step":
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"epoch": 0.
|
| 96 |
-
"learning_rate":
|
| 97 |
-
"loss": 2.
|
| 98 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 0.49,
|
| 102 |
-
"learning_rate":
|
| 103 |
-
"loss": 2.
|
| 104 |
-
"step":
|
| 105 |
},
|
| 106 |
{
|
| 107 |
"epoch": 0.5,
|
| 108 |
-
"eval_accuracy": 0.
|
| 109 |
-
"eval_loss": 2.
|
| 110 |
-
"eval_runtime":
|
| 111 |
-
"eval_samples_per_second": 3.
|
| 112 |
-
"eval_steps_per_second": 3.
|
| 113 |
-
"step":
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"epoch": 0.
|
| 117 |
-
"learning_rate":
|
| 118 |
-
"loss": 2.
|
| 119 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 120 |
},
|
| 121 |
{
|
| 122 |
"epoch": 0.55,
|
| 123 |
-
"learning_rate":
|
| 124 |
-
"loss": 2.
|
| 125 |
-
"step":
|
| 126 |
},
|
| 127 |
{
|
| 128 |
-
"epoch": 0.
|
| 129 |
-
"learning_rate":
|
| 130 |
-
"loss": 2.
|
| 131 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 132 |
},
|
| 133 |
{
|
| 134 |
"epoch": 0.61,
|
| 135 |
-
"learning_rate":
|
| 136 |
-
"loss": 2.
|
| 137 |
-
"step":
|
| 138 |
},
|
| 139 |
{
|
| 140 |
-
"epoch": 0.
|
| 141 |
-
"learning_rate":
|
| 142 |
-
"loss": 2.
|
| 143 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 144 |
},
|
| 145 |
{
|
| 146 |
"epoch": 0.68,
|
| 147 |
-
"learning_rate":
|
| 148 |
-
"loss": 2.
|
| 149 |
-
"step":
|
| 150 |
},
|
| 151 |
{
|
| 152 |
-
"epoch": 0.
|
| 153 |
-
"learning_rate":
|
| 154 |
-
"loss": 2.
|
| 155 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 156 |
},
|
| 157 |
{
|
| 158 |
"epoch": 0.74,
|
| 159 |
-
"learning_rate":
|
| 160 |
-
"loss": 2.
|
| 161 |
-
"step":
|
| 162 |
},
|
| 163 |
{
|
| 164 |
-
"epoch": 0.
|
| 165 |
-
"learning_rate":
|
| 166 |
-
"loss": 2.
|
| 167 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 168 |
},
|
| 169 |
{
|
| 170 |
"epoch": 0.8,
|
| 171 |
-
"learning_rate":
|
| 172 |
-
"loss": 2.
|
| 173 |
-
"step":
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"epoch": 0.
|
| 177 |
-
"learning_rate":
|
| 178 |
-
"loss": 2.
|
| 179 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 0.86,
|
| 183 |
-
"learning_rate":
|
| 184 |
-
"loss": 2.
|
| 185 |
-
"step":
|
| 186 |
},
|
| 187 |
{
|
| 188 |
-
"epoch": 0.
|
| 189 |
-
"learning_rate":
|
| 190 |
-
"loss": 2.
|
| 191 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 192 |
},
|
| 193 |
{
|
| 194 |
"epoch": 0.92,
|
| 195 |
-
"learning_rate":
|
| 196 |
-
"loss": 2.
|
| 197 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 198 |
},
|
| 199 |
{
|
| 200 |
-
"epoch": 0.
|
| 201 |
-
"learning_rate":
|
| 202 |
-
"loss": 2.
|
| 203 |
-
"step":
|
| 204 |
},
|
| 205 |
{
|
| 206 |
"epoch": 0.98,
|
| 207 |
-
"learning_rate":
|
| 208 |
-
"loss": 2.
|
| 209 |
-
"step":
|
| 210 |
},
|
| 211 |
{
|
| 212 |
"epoch": 1.0,
|
| 213 |
-
"
|
| 214 |
-
"
|
| 215 |
-
"
|
| 216 |
-
|
| 217 |
-
|
| 218 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 219 |
},
|
| 220 |
{
|
| 221 |
-
"epoch": 1.
|
| 222 |
-
"learning_rate":
|
| 223 |
-
"loss": 1.
|
| 224 |
-
"step":
|
| 225 |
},
|
| 226 |
{
|
| 227 |
"epoch": 1.05,
|
| 228 |
-
"learning_rate":
|
| 229 |
-
"loss": 1.
|
| 230 |
-
"step":
|
| 231 |
},
|
| 232 |
{
|
| 233 |
-
"epoch": 1.
|
| 234 |
-
"learning_rate":
|
| 235 |
-
"loss":
|
| 236 |
-
"step":
|
| 237 |
},
|
| 238 |
{
|
| 239 |
-
"epoch": 1.
|
| 240 |
-
"learning_rate":
|
| 241 |
-
"loss": 1.
|
| 242 |
-
"step":
|
| 243 |
},
|
| 244 |
{
|
| 245 |
-
"epoch": 1.
|
| 246 |
-
"learning_rate":
|
| 247 |
-
"loss":
|
| 248 |
-
"step":
|
| 249 |
},
|
| 250 |
{
|
| 251 |
-
"epoch": 1.
|
| 252 |
-
"learning_rate":
|
| 253 |
-
"loss": 1.
|
| 254 |
-
"step":
|
| 255 |
},
|
| 256 |
{
|
| 257 |
-
"epoch": 1.
|
| 258 |
-
"learning_rate":
|
| 259 |
-
"loss": 1.
|
| 260 |
-
"step":
|
| 261 |
},
|
| 262 |
{
|
| 263 |
-
"epoch": 1.
|
| 264 |
-
"learning_rate":
|
| 265 |
-
"loss": 1.
|
| 266 |
-
"step":
|
| 267 |
},
|
| 268 |
{
|
| 269 |
-
"epoch": 1.
|
| 270 |
-
"learning_rate":
|
| 271 |
-
"loss": 1.
|
| 272 |
-
"step":
|
| 273 |
},
|
| 274 |
{
|
| 275 |
-
"epoch": 1.
|
| 276 |
-
"learning_rate":
|
| 277 |
-
"loss": 1.
|
| 278 |
-
"step":
|
| 279 |
},
|
| 280 |
{
|
| 281 |
-
"epoch": 1.
|
| 282 |
-
"learning_rate":
|
| 283 |
-
"loss": 1.
|
| 284 |
-
"step":
|
| 285 |
},
|
| 286 |
{
|
| 287 |
-
"epoch": 1.
|
| 288 |
-
"learning_rate":
|
| 289 |
-
"loss": 1.
|
| 290 |
-
"step":
|
| 291 |
},
|
| 292 |
{
|
| 293 |
-
"epoch": 1.
|
| 294 |
-
"learning_rate":
|
| 295 |
-
"loss": 1.
|
| 296 |
-
"step":
|
| 297 |
},
|
| 298 |
{
|
| 299 |
-
"epoch": 1.
|
| 300 |
-
"learning_rate":
|
| 301 |
-
"loss": 1.
|
| 302 |
-
"step":
|
| 303 |
},
|
| 304 |
{
|
| 305 |
-
"epoch": 1.
|
| 306 |
-
"learning_rate":
|
| 307 |
-
"loss": 1.
|
| 308 |
-
"step":
|
| 309 |
},
|
| 310 |
{
|
| 311 |
-
"epoch": 1.
|
| 312 |
-
"learning_rate":
|
| 313 |
-
"loss": 1.
|
| 314 |
-
"step":
|
| 315 |
},
|
| 316 |
{
|
| 317 |
-
"epoch": 1.
|
| 318 |
-
"
|
| 319 |
-
"
|
| 320 |
-
"
|
| 321 |
-
"eval_samples_per_second": 3.303,
|
| 322 |
-
"eval_steps_per_second": 3.303,
|
| 323 |
-
"step": 489
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"epoch": 1.
|
| 327 |
-
"learning_rate":
|
| 328 |
-
"loss": 1.
|
| 329 |
-
"step":
|
| 330 |
},
|
| 331 |
{
|
| 332 |
-
"epoch": 1.
|
| 333 |
-
"learning_rate":
|
| 334 |
-
"loss": 1.
|
| 335 |
-
"step":
|
| 336 |
},
|
| 337 |
{
|
| 338 |
-
"epoch": 1.
|
| 339 |
-
"learning_rate":
|
| 340 |
-
"loss": 1.
|
| 341 |
-
"step":
|
| 342 |
},
|
| 343 |
{
|
| 344 |
-
"epoch": 1.
|
| 345 |
-
"learning_rate":
|
| 346 |
-
"loss": 1.
|
| 347 |
-
"step":
|
| 348 |
},
|
| 349 |
{
|
| 350 |
-
"epoch": 1.
|
| 351 |
-
"learning_rate":
|
| 352 |
-
"loss": 1.
|
| 353 |
-
"step":
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"epoch": 1.
|
| 357 |
-
"learning_rate":
|
| 358 |
-
"loss": 1.
|
| 359 |
-
"step":
|
| 360 |
},
|
| 361 |
{
|
| 362 |
-
"epoch": 1.
|
| 363 |
-
"learning_rate":
|
| 364 |
-
"loss": 1.
|
| 365 |
-
"step":
|
| 366 |
},
|
| 367 |
{
|
| 368 |
-
"epoch": 1.
|
| 369 |
-
"
|
| 370 |
-
"
|
| 371 |
-
"
|
|
|
|
|
|
|
|
|
|
| 372 |
},
|
| 373 |
{
|
| 374 |
-
"epoch": 1.
|
| 375 |
-
"learning_rate":
|
| 376 |
-
"loss": 1.
|
| 377 |
-
"step":
|
| 378 |
},
|
| 379 |
{
|
| 380 |
-
"epoch": 1.
|
| 381 |
-
"learning_rate":
|
| 382 |
-
"loss": 1.
|
| 383 |
-
"step":
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"epoch": 1.
|
| 387 |
-
"learning_rate":
|
| 388 |
-
"loss": 1.
|
| 389 |
-
"step":
|
| 390 |
},
|
| 391 |
{
|
| 392 |
-
"epoch": 1.
|
| 393 |
-
"learning_rate":
|
| 394 |
-
"loss": 1.
|
| 395 |
-
"step":
|
| 396 |
},
|
| 397 |
{
|
| 398 |
-
"epoch": 1.
|
| 399 |
-
"learning_rate":
|
| 400 |
-
"loss": 1.
|
| 401 |
-
"step":
|
| 402 |
},
|
| 403 |
{
|
| 404 |
-
"epoch": 1.
|
| 405 |
-
"learning_rate":
|
| 406 |
-
"loss": 1.
|
| 407 |
-
"step":
|
| 408 |
},
|
| 409 |
{
|
| 410 |
-
"epoch": 1.
|
| 411 |
-
"learning_rate":
|
| 412 |
-
"loss": 1.
|
| 413 |
-
"step":
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"epoch": 1.
|
| 417 |
-
"learning_rate":
|
| 418 |
-
"loss": 1.
|
| 419 |
-
"step":
|
| 420 |
},
|
| 421 |
{
|
| 422 |
-
"epoch":
|
| 423 |
-
"learning_rate":
|
| 424 |
-
"loss": 1.
|
| 425 |
-
"step":
|
| 426 |
},
|
| 427 |
{
|
| 428 |
-
"epoch":
|
| 429 |
-
"
|
| 430 |
-
"
|
| 431 |
-
"
|
| 432 |
-
"eval_samples_per_second": 3.304,
|
| 433 |
-
"eval_steps_per_second": 3.304,
|
| 434 |
-
"step": 652
|
| 435 |
},
|
| 436 |
{
|
| 437 |
-
"epoch":
|
| 438 |
-
"learning_rate":
|
| 439 |
-
"loss": 1.
|
| 440 |
-
"step":
|
| 441 |
},
|
| 442 |
{
|
| 443 |
-
"epoch":
|
| 444 |
-
"learning_rate":
|
| 445 |
-
"loss": 1.
|
| 446 |
-
"step":
|
| 447 |
},
|
| 448 |
{
|
| 449 |
-
"epoch":
|
| 450 |
-
"learning_rate":
|
| 451 |
-
"loss": 1.
|
| 452 |
-
"step":
|
| 453 |
},
|
| 454 |
{
|
| 455 |
-
"epoch":
|
| 456 |
-
"learning_rate":
|
| 457 |
-
"loss": 1.
|
| 458 |
-
"step":
|
| 459 |
},
|
| 460 |
{
|
| 461 |
-
"epoch":
|
| 462 |
-
"learning_rate":
|
| 463 |
-
"loss": 1.
|
| 464 |
-
"step":
|
| 465 |
},
|
| 466 |
{
|
| 467 |
-
"epoch":
|
| 468 |
-
"learning_rate":
|
| 469 |
-
"loss": 1.
|
| 470 |
-
"step":
|
| 471 |
},
|
| 472 |
{
|
| 473 |
-
"epoch":
|
| 474 |
-
"learning_rate":
|
| 475 |
-
"loss": 1.
|
| 476 |
-
"step":
|
| 477 |
},
|
| 478 |
{
|
| 479 |
-
"epoch":
|
| 480 |
-
"learning_rate":
|
| 481 |
-
"loss": 1.
|
| 482 |
-
"step":
|
| 483 |
},
|
| 484 |
{
|
| 485 |
-
"epoch":
|
| 486 |
-
"learning_rate":
|
| 487 |
-
"loss": 1.
|
| 488 |
-
"step":
|
| 489 |
},
|
| 490 |
{
|
| 491 |
-
"epoch":
|
| 492 |
-
"learning_rate":
|
| 493 |
-
"loss": 1.
|
| 494 |
-
"step":
|
| 495 |
},
|
| 496 |
{
|
| 497 |
-
"epoch":
|
| 498 |
-
"learning_rate":
|
| 499 |
-
"loss": 1.
|
| 500 |
-
"step":
|
| 501 |
},
|
| 502 |
{
|
| 503 |
-
"epoch":
|
| 504 |
-
"learning_rate":
|
| 505 |
-
"loss": 1.
|
| 506 |
-
"step":
|
| 507 |
},
|
| 508 |
{
|
| 509 |
-
"epoch":
|
| 510 |
-
"learning_rate":
|
| 511 |
-
"loss": 1.
|
| 512 |
-
"step":
|
| 513 |
},
|
| 514 |
{
|
| 515 |
-
"epoch":
|
| 516 |
-
"learning_rate":
|
| 517 |
-
"loss": 1.
|
| 518 |
-
"step":
|
| 519 |
},
|
| 520 |
{
|
| 521 |
-
"epoch":
|
| 522 |
-
"learning_rate":
|
| 523 |
-
"loss": 1.
|
| 524 |
-
"step":
|
| 525 |
},
|
| 526 |
{
|
| 527 |
-
"epoch":
|
| 528 |
-
"learning_rate":
|
| 529 |
-
"loss": 1.
|
| 530 |
-
"step":
|
| 531 |
},
|
| 532 |
{
|
| 533 |
-
"epoch":
|
| 534 |
-
"eval_accuracy": 0.
|
| 535 |
-
"eval_loss": 2.
|
| 536 |
-
"eval_runtime":
|
| 537 |
-
"eval_samples_per_second": 3.
|
| 538 |
-
"eval_steps_per_second": 3.
|
| 539 |
-
"step":
|
| 540 |
},
|
| 541 |
{
|
| 542 |
-
"epoch": 2.
|
| 543 |
-
"learning_rate":
|
| 544 |
-
"loss": 1.
|
| 545 |
-
"step":
|
| 546 |
},
|
| 547 |
{
|
| 548 |
-
"epoch": 2.
|
| 549 |
-
"learning_rate":
|
| 550 |
-
"loss": 1.
|
| 551 |
-
"step":
|
| 552 |
},
|
| 553 |
{
|
| 554 |
-
"epoch": 2.
|
| 555 |
-
"learning_rate": 1.
|
| 556 |
-
"loss": 1.
|
| 557 |
-
"step":
|
| 558 |
},
|
| 559 |
{
|
| 560 |
-
"epoch": 2.
|
| 561 |
-
"learning_rate": 1.
|
| 562 |
-
"loss": 1.
|
| 563 |
-
"step":
|
| 564 |
},
|
| 565 |
{
|
| 566 |
-
"epoch": 2.
|
| 567 |
-
"learning_rate": 1.
|
| 568 |
-
"loss": 1.
|
| 569 |
-
"step":
|
| 570 |
},
|
| 571 |
{
|
| 572 |
-
"epoch": 2.
|
| 573 |
-
"learning_rate":
|
| 574 |
-
"loss": 1.
|
| 575 |
-
"step":
|
| 576 |
},
|
| 577 |
{
|
| 578 |
-
"epoch": 2.
|
| 579 |
-
"learning_rate":
|
| 580 |
-
"loss": 1.
|
| 581 |
-
"step":
|
| 582 |
},
|
| 583 |
{
|
| 584 |
-
"epoch": 2.
|
| 585 |
-
"learning_rate":
|
| 586 |
-
"loss": 1.
|
| 587 |
-
"step":
|
| 588 |
},
|
| 589 |
{
|
| 590 |
-
"epoch": 2.
|
| 591 |
-
"learning_rate":
|
| 592 |
-
"loss": 1.
|
| 593 |
-
"step":
|
| 594 |
},
|
| 595 |
{
|
| 596 |
-
"epoch": 2.
|
| 597 |
-
"learning_rate":
|
| 598 |
-
"loss": 1.
|
| 599 |
-
"step":
|
| 600 |
},
|
| 601 |
{
|
| 602 |
-
"epoch": 2.
|
| 603 |
-
"learning_rate":
|
| 604 |
-
"loss": 1.
|
| 605 |
-
"step":
|
| 606 |
},
|
| 607 |
{
|
| 608 |
-
"epoch": 2.
|
| 609 |
-
"learning_rate":
|
| 610 |
-
"loss": 1.
|
| 611 |
-
"step":
|
| 612 |
},
|
| 613 |
{
|
| 614 |
-
"epoch": 2.
|
| 615 |
-
"learning_rate":
|
| 616 |
-
"loss": 1.
|
| 617 |
-
"step":
|
| 618 |
},
|
| 619 |
{
|
| 620 |
-
"epoch": 2.
|
| 621 |
-
"learning_rate":
|
| 622 |
-
"loss": 1.
|
| 623 |
-
"step":
|
| 624 |
},
|
| 625 |
{
|
| 626 |
-
"epoch": 2.
|
| 627 |
-
"learning_rate":
|
| 628 |
-
"loss": 1.
|
| 629 |
-
"step":
|
| 630 |
},
|
| 631 |
{
|
| 632 |
-
"epoch": 2.
|
| 633 |
-
"learning_rate":
|
| 634 |
-
"loss": 1.
|
| 635 |
-
"step":
|
| 636 |
},
|
| 637 |
{
|
| 638 |
-
"epoch":
|
| 639 |
-
"
|
| 640 |
-
"
|
| 641 |
-
"
|
| 642 |
-
"eval_samples_per_second": 3.294,
|
| 643 |
-
"eval_steps_per_second": 3.294,
|
| 644 |
-
"step": 978
|
| 645 |
},
|
| 646 |
{
|
| 647 |
-
"epoch":
|
| 648 |
-
"learning_rate":
|
| 649 |
-
"loss": 1.
|
| 650 |
-
"step":
|
| 651 |
},
|
| 652 |
{
|
| 653 |
-
"epoch":
|
| 654 |
-
"learning_rate":
|
| 655 |
-
"loss": 1.
|
| 656 |
-
"step":
|
| 657 |
},
|
| 658 |
{
|
| 659 |
-
"epoch":
|
| 660 |
-
"learning_rate":
|
| 661 |
-
"loss": 1.
|
| 662 |
-
"step":
|
| 663 |
},
|
| 664 |
{
|
| 665 |
-
"epoch":
|
| 666 |
-
"learning_rate":
|
| 667 |
-
"loss": 1.
|
| 668 |
-
"step":
|
| 669 |
},
|
| 670 |
{
|
| 671 |
-
"epoch":
|
| 672 |
-
"learning_rate":
|
| 673 |
-
"loss": 1.
|
| 674 |
-
"step":
|
| 675 |
},
|
| 676 |
{
|
| 677 |
-
"epoch":
|
| 678 |
-
"learning_rate":
|
| 679 |
-
"loss": 1.
|
| 680 |
-
"step":
|
| 681 |
},
|
| 682 |
{
|
| 683 |
-
"epoch":
|
| 684 |
-
"learning_rate":
|
| 685 |
-
"loss": 1.
|
| 686 |
-
"step":
|
| 687 |
},
|
| 688 |
{
|
| 689 |
-
"epoch":
|
| 690 |
-
"learning_rate":
|
| 691 |
-
"loss": 1.
|
| 692 |
-
"step":
|
| 693 |
},
|
| 694 |
{
|
| 695 |
-
"epoch":
|
| 696 |
-
"
|
| 697 |
-
"
|
| 698 |
-
"
|
|
|
|
|
|
|
|
|
|
| 699 |
},
|
| 700 |
{
|
| 701 |
-
"epoch":
|
| 702 |
-
"learning_rate":
|
| 703 |
-
"loss": 1.
|
| 704 |
-
"step":
|
| 705 |
},
|
| 706 |
{
|
| 707 |
-
"epoch":
|
| 708 |
-
"learning_rate":
|
| 709 |
-
"loss": 1.
|
| 710 |
-
"step":
|
| 711 |
},
|
| 712 |
{
|
| 713 |
-
"epoch":
|
| 714 |
-
"learning_rate":
|
| 715 |
-
"loss": 1.
|
| 716 |
-
"step":
|
| 717 |
},
|
| 718 |
{
|
| 719 |
-
"epoch":
|
| 720 |
-
"learning_rate":
|
| 721 |
-
"loss": 1.
|
| 722 |
-
"step":
|
| 723 |
},
|
| 724 |
{
|
| 725 |
-
"epoch":
|
| 726 |
-
"learning_rate":
|
| 727 |
-
"loss": 1.
|
| 728 |
-
"step":
|
| 729 |
},
|
| 730 |
{
|
| 731 |
-
"epoch":
|
| 732 |
-
"learning_rate":
|
| 733 |
-
"loss": 1.
|
| 734 |
-
"step":
|
| 735 |
},
|
| 736 |
{
|
| 737 |
-
"epoch":
|
| 738 |
-
"learning_rate":
|
| 739 |
-
"loss": 1.
|
| 740 |
-
"step":
|
| 741 |
},
|
| 742 |
{
|
| 743 |
-
"epoch":
|
| 744 |
-
"learning_rate":
|
| 745 |
-
"loss": 1.
|
| 746 |
-
"step":
|
| 747 |
},
|
| 748 |
{
|
| 749 |
-
"epoch":
|
| 750 |
-
"
|
| 751 |
-
"
|
| 752 |
-
"
|
| 753 |
-
"eval_samples_per_second": 3.305,
|
| 754 |
-
"eval_steps_per_second": 3.305,
|
| 755 |
-
"step": 1141
|
| 756 |
},
|
| 757 |
{
|
| 758 |
-
"epoch":
|
| 759 |
-
"learning_rate":
|
| 760 |
-
"loss": 1.
|
| 761 |
-
"step":
|
| 762 |
},
|
| 763 |
{
|
| 764 |
-
"epoch":
|
| 765 |
-
"learning_rate":
|
| 766 |
-
"loss": 1.
|
| 767 |
-
"step":
|
| 768 |
},
|
| 769 |
{
|
| 770 |
-
"epoch":
|
| 771 |
-
"learning_rate":
|
| 772 |
-
"loss": 1.
|
| 773 |
-
"step":
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"epoch":
|
| 777 |
-
"learning_rate":
|
| 778 |
-
"loss": 1.
|
| 779 |
-
"step":
|
| 780 |
},
|
| 781 |
{
|
| 782 |
-
"epoch":
|
| 783 |
-
"learning_rate":
|
| 784 |
-
"loss": 1.
|
| 785 |
-
"step":
|
| 786 |
},
|
| 787 |
{
|
| 788 |
-
"epoch":
|
| 789 |
-
"learning_rate":
|
| 790 |
-
"loss": 1.
|
| 791 |
-
"step":
|
| 792 |
},
|
| 793 |
{
|
| 794 |
-
"epoch":
|
| 795 |
-
"learning_rate":
|
| 796 |
-
"loss": 1.
|
| 797 |
-
"step":
|
| 798 |
},
|
| 799 |
{
|
| 800 |
-
"epoch":
|
| 801 |
-
"learning_rate":
|
| 802 |
-
"loss": 1.
|
| 803 |
-
"step":
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"epoch":
|
| 807 |
-
"learning_rate":
|
| 808 |
-
"loss": 1.
|
| 809 |
-
"step":
|
| 810 |
},
|
| 811 |
{
|
| 812 |
-
"epoch":
|
| 813 |
-
"learning_rate":
|
| 814 |
-
"loss": 1.
|
| 815 |
-
"step":
|
| 816 |
},
|
| 817 |
{
|
| 818 |
-
"epoch":
|
| 819 |
-
"learning_rate":
|
| 820 |
-
"loss": 1.
|
| 821 |
-
"step":
|
| 822 |
},
|
| 823 |
{
|
| 824 |
-
"epoch":
|
| 825 |
-
"learning_rate": 1.
|
| 826 |
-
"loss": 1.
|
| 827 |
-
"step":
|
| 828 |
},
|
| 829 |
{
|
| 830 |
-
"epoch":
|
| 831 |
-
"learning_rate":
|
| 832 |
-
"loss": 1.
|
| 833 |
-
"step":
|
| 834 |
},
|
| 835 |
{
|
| 836 |
-
"epoch":
|
| 837 |
-
"learning_rate":
|
| 838 |
-
"loss": 1.
|
| 839 |
-
"step":
|
| 840 |
},
|
| 841 |
{
|
| 842 |
-
"epoch":
|
| 843 |
-
"learning_rate": 4.
|
| 844 |
-
"loss": 1.
|
| 845 |
-
"step":
|
| 846 |
},
|
| 847 |
{
|
| 848 |
-
"epoch":
|
| 849 |
-
"learning_rate":
|
| 850 |
-
"loss": 1.
|
| 851 |
-
"step":
|
| 852 |
},
|
| 853 |
{
|
| 854 |
-
"epoch":
|
| 855 |
-
"
|
| 856 |
-
"
|
| 857 |
-
"
|
| 858 |
-
|
| 859 |
-
|
| 860 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 861 |
}
|
| 862 |
],
|
| 863 |
-
"max_steps":
|
| 864 |
-
"num_train_epochs":
|
| 865 |
-
"total_flos": 4.
|
| 866 |
"trial_name": null,
|
| 867 |
"trial_params": null
|
| 868 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.6870071488415761,
|
| 3 |
+
"best_model_checkpoint": "opt-2.7b-realtime-chat-v2/checkpoint-204",
|
| 4 |
+
"epoch": 2.9874151475859962,
|
| 5 |
+
"global_step": 306,
|
| 6 |
"is_hyper_param_search": false,
|
| 7 |
"is_local_process_zero": true,
|
| 8 |
"is_world_process_zero": true,
|
| 9 |
"log_history": [
|
| 10 |
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 1.935483870967742e-06,
|
| 13 |
+
"loss": 2.6432,
|
| 14 |
+
"step": 2
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 3.870967741935484e-06,
|
| 19 |
+
"loss": 2.6152,
|
| 20 |
+
"step": 4
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"epoch": 0.06,
|
| 24 |
+
"learning_rate": 5.8064516129032256e-06,
|
| 25 |
+
"loss": 2.5393,
|
| 26 |
+
"step": 6
|
| 27 |
},
|
| 28 |
{
|
| 29 |
+
"epoch": 0.08,
|
| 30 |
+
"learning_rate": 7.741935483870968e-06,
|
| 31 |
+
"loss": 2.4402,
|
| 32 |
+
"step": 8
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.1,
|
| 36 |
+
"learning_rate": 9.67741935483871e-06,
|
| 37 |
+
"loss": 2.3797,
|
| 38 |
+
"step": 10
|
| 39 |
},
|
| 40 |
{
|
| 41 |
"epoch": 0.12,
|
| 42 |
+
"learning_rate": 1.1612903225806451e-05,
|
| 43 |
+
"loss": 2.3533,
|
| 44 |
+
"step": 12
|
| 45 |
},
|
| 46 |
{
|
| 47 |
+
"epoch": 0.14,
|
| 48 |
+
"learning_rate": 1.3548387096774194e-05,
|
| 49 |
+
"loss": 2.307,
|
| 50 |
+
"step": 14
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.16,
|
| 54 |
+
"learning_rate": 1.5483870967741936e-05,
|
| 55 |
+
"loss": 2.266,
|
| 56 |
+
"step": 16
|
| 57 |
},
|
| 58 |
{
|
| 59 |
"epoch": 0.18,
|
| 60 |
+
"learning_rate": 1.741935483870968e-05,
|
| 61 |
+
"loss": 2.2859,
|
| 62 |
+
"step": 18
|
| 63 |
},
|
| 64 |
{
|
| 65 |
+
"epoch": 0.2,
|
| 66 |
+
"learning_rate": 1.935483870967742e-05,
|
| 67 |
+
"loss": 2.2434,
|
| 68 |
+
"step": 20
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.21,
|
| 72 |
+
"learning_rate": 2.1290322580645163e-05,
|
| 73 |
+
"loss": 2.2224,
|
| 74 |
+
"step": 22
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.23,
|
| 78 |
+
"learning_rate": 2.3225806451612902e-05,
|
| 79 |
+
"loss": 2.2518,
|
| 80 |
+
"step": 24
|
| 81 |
},
|
| 82 |
{
|
| 83 |
"epoch": 0.25,
|
| 84 |
+
"learning_rate": 2.5161290322580648e-05,
|
| 85 |
+
"loss": 2.232,
|
| 86 |
+
"step": 26
|
| 87 |
},
|
| 88 |
{
|
| 89 |
+
"epoch": 0.27,
|
| 90 |
+
"learning_rate": 2.7096774193548387e-05,
|
| 91 |
+
"loss": 2.2121,
|
| 92 |
+
"step": 28
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.29,
|
| 96 |
+
"learning_rate": 2.903225806451613e-05,
|
| 97 |
+
"loss": 2.2042,
|
| 98 |
+
"step": 30
|
| 99 |
},
|
| 100 |
{
|
| 101 |
"epoch": 0.31,
|
| 102 |
+
"learning_rate": 2.9890909090909092e-05,
|
| 103 |
+
"loss": 2.1655,
|
| 104 |
+
"step": 32
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"epoch": 0.33,
|
| 108 |
+
"learning_rate": 2.9672727272727274e-05,
|
| 109 |
+
"loss": 2.197,
|
| 110 |
+
"step": 34
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.35,
|
| 114 |
+
"learning_rate": 2.9454545454545456e-05,
|
| 115 |
+
"loss": 2.2383,
|
| 116 |
+
"step": 36
|
| 117 |
},
|
| 118 |
{
|
| 119 |
"epoch": 0.37,
|
| 120 |
+
"learning_rate": 2.9236363636363635e-05,
|
| 121 |
+
"loss": 2.1827,
|
| 122 |
+
"step": 38
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"epoch": 0.39,
|
| 126 |
+
"learning_rate": 2.901818181818182e-05,
|
| 127 |
+
"loss": 2.1667,
|
| 128 |
+
"step": 40
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.41,
|
| 132 |
+
"learning_rate": 2.88e-05,
|
| 133 |
+
"loss": 2.1688,
|
| 134 |
+
"step": 42
|
| 135 |
},
|
| 136 |
{
|
| 137 |
"epoch": 0.43,
|
| 138 |
+
"learning_rate": 2.8581818181818184e-05,
|
| 139 |
+
"loss": 2.1722,
|
| 140 |
+
"step": 44
|
| 141 |
},
|
| 142 |
{
|
| 143 |
+
"epoch": 0.45,
|
| 144 |
+
"learning_rate": 2.8363636363636363e-05,
|
| 145 |
+
"loss": 2.1771,
|
| 146 |
+
"step": 46
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.47,
|
| 150 |
+
"learning_rate": 2.8145454545454548e-05,
|
| 151 |
+
"loss": 2.1681,
|
| 152 |
+
"step": 48
|
| 153 |
},
|
| 154 |
{
|
| 155 |
"epoch": 0.49,
|
| 156 |
+
"learning_rate": 2.7927272727272727e-05,
|
| 157 |
+
"loss": 2.0974,
|
| 158 |
+
"step": 50
|
| 159 |
},
|
| 160 |
{
|
| 161 |
"epoch": 0.5,
|
| 162 |
+
"eval_accuracy": 0.6825691202041083,
|
| 163 |
+
"eval_loss": 2.1267497539520264,
|
| 164 |
+
"eval_runtime": 510.4375,
|
| 165 |
+
"eval_samples_per_second": 3.519,
|
| 166 |
+
"eval_steps_per_second": 3.519,
|
| 167 |
+
"step": 51
|
| 168 |
},
|
| 169 |
{
|
| 170 |
+
"epoch": 0.51,
|
| 171 |
+
"learning_rate": 2.770909090909091e-05,
|
| 172 |
+
"loss": 2.1322,
|
| 173 |
+
"step": 52
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.53,
|
| 177 |
+
"learning_rate": 2.749090909090909e-05,
|
| 178 |
+
"loss": 2.1157,
|
| 179 |
+
"step": 54
|
| 180 |
},
|
| 181 |
{
|
| 182 |
"epoch": 0.55,
|
| 183 |
+
"learning_rate": 2.7272727272727273e-05,
|
| 184 |
+
"loss": 2.1325,
|
| 185 |
+
"step": 56
|
| 186 |
},
|
| 187 |
{
|
| 188 |
+
"epoch": 0.57,
|
| 189 |
+
"learning_rate": 2.7054545454545455e-05,
|
| 190 |
+
"loss": 2.1346,
|
| 191 |
+
"step": 58
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.59,
|
| 195 |
+
"learning_rate": 2.6836363636363637e-05,
|
| 196 |
+
"loss": 2.143,
|
| 197 |
+
"step": 60
|
| 198 |
},
|
| 199 |
{
|
| 200 |
"epoch": 0.61,
|
| 201 |
+
"learning_rate": 2.661818181818182e-05,
|
| 202 |
+
"loss": 2.143,
|
| 203 |
+
"step": 62
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"epoch": 0.62,
|
| 207 |
+
"learning_rate": 2.64e-05,
|
| 208 |
+
"loss": 2.1353,
|
| 209 |
+
"step": 64
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.64,
|
| 213 |
+
"learning_rate": 2.618181818181818e-05,
|
| 214 |
+
"loss": 2.1296,
|
| 215 |
+
"step": 66
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.66,
|
| 219 |
+
"learning_rate": 2.5963636363636365e-05,
|
| 220 |
+
"loss": 2.117,
|
| 221 |
+
"step": 68
|
| 222 |
},
|
| 223 |
{
|
| 224 |
"epoch": 0.68,
|
| 225 |
+
"learning_rate": 2.5745454545454544e-05,
|
| 226 |
+
"loss": 2.1093,
|
| 227 |
+
"step": 70
|
| 228 |
},
|
| 229 |
{
|
| 230 |
+
"epoch": 0.7,
|
| 231 |
+
"learning_rate": 2.552727272727273e-05,
|
| 232 |
+
"loss": 2.1206,
|
| 233 |
+
"step": 72
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.72,
|
| 237 |
+
"learning_rate": 2.5309090909090908e-05,
|
| 238 |
+
"loss": 2.1308,
|
| 239 |
+
"step": 74
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"epoch": 0.74,
|
| 243 |
+
"learning_rate": 2.509090909090909e-05,
|
| 244 |
+
"loss": 2.1258,
|
| 245 |
+
"step": 76
|
| 246 |
},
|
| 247 |
{
|
| 248 |
+
"epoch": 0.76,
|
| 249 |
+
"learning_rate": 2.4872727272727272e-05,
|
| 250 |
+
"loss": 2.1113,
|
| 251 |
+
"step": 78
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.78,
|
| 255 |
+
"learning_rate": 2.4654545454545454e-05,
|
| 256 |
+
"loss": 2.0913,
|
| 257 |
+
"step": 80
|
| 258 |
},
|
| 259 |
{
|
| 260 |
"epoch": 0.8,
|
| 261 |
+
"learning_rate": 2.4436363636363636e-05,
|
| 262 |
+
"loss": 2.1067,
|
| 263 |
+
"step": 82
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"epoch": 0.82,
|
| 267 |
+
"learning_rate": 2.421818181818182e-05,
|
| 268 |
+
"loss": 2.1034,
|
| 269 |
+
"step": 84
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.84,
|
| 273 |
+
"learning_rate": 2.4e-05,
|
| 274 |
+
"loss": 2.0804,
|
| 275 |
+
"step": 86
|
| 276 |
},
|
| 277 |
{
|
| 278 |
"epoch": 0.86,
|
| 279 |
+
"learning_rate": 2.3781818181818183e-05,
|
| 280 |
+
"loss": 2.1228,
|
| 281 |
+
"step": 88
|
| 282 |
},
|
| 283 |
{
|
| 284 |
+
"epoch": 0.88,
|
| 285 |
+
"learning_rate": 2.356363636363636e-05,
|
| 286 |
+
"loss": 2.1044,
|
| 287 |
+
"step": 90
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.9,
|
| 291 |
+
"learning_rate": 2.3345454545454547e-05,
|
| 292 |
+
"loss": 2.1045,
|
| 293 |
+
"step": 92
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"epoch": 0.92,
|
| 297 |
+
"learning_rate": 2.3127272727272725e-05,
|
| 298 |
+
"loss": 2.0758,
|
| 299 |
+
"step": 94
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.94,
|
| 303 |
+
"learning_rate": 2.290909090909091e-05,
|
| 304 |
+
"loss": 2.1185,
|
| 305 |
+
"step": 96
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"epoch": 0.96,
|
| 309 |
+
"learning_rate": 2.269090909090909e-05,
|
| 310 |
+
"loss": 2.0844,
|
| 311 |
+
"step": 98
|
| 312 |
},
|
| 313 |
{
|
| 314 |
"epoch": 0.98,
|
| 315 |
+
"learning_rate": 2.2472727272727275e-05,
|
| 316 |
+
"loss": 2.0817,
|
| 317 |
+
"step": 100
|
| 318 |
},
|
| 319 |
{
|
| 320 |
"epoch": 1.0,
|
| 321 |
+
"learning_rate": 2.2254545454545454e-05,
|
| 322 |
+
"loss": 2.0842,
|
| 323 |
+
"step": 102
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.0,
|
| 327 |
+
"eval_accuracy": 0.6858646182863831,
|
| 328 |
+
"eval_loss": 2.096801519393921,
|
| 329 |
+
"eval_runtime": 510.9612,
|
| 330 |
+
"eval_samples_per_second": 3.515,
|
| 331 |
+
"eval_steps_per_second": 3.515,
|
| 332 |
+
"step": 102
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.02,
|
| 336 |
+
"learning_rate": 2.2036363636363636e-05,
|
| 337 |
+
"loss": 2.0166,
|
| 338 |
+
"step": 104
|
| 339 |
},
|
| 340 |
{
|
| 341 |
+
"epoch": 1.03,
|
| 342 |
+
"learning_rate": 2.1818181818181818e-05,
|
| 343 |
+
"loss": 1.9987,
|
| 344 |
+
"step": 106
|
| 345 |
},
|
| 346 |
{
|
| 347 |
"epoch": 1.05,
|
| 348 |
+
"learning_rate": 2.16e-05,
|
| 349 |
+
"loss": 1.9678,
|
| 350 |
+
"step": 108
|
| 351 |
},
|
| 352 |
{
|
| 353 |
+
"epoch": 1.07,
|
| 354 |
+
"learning_rate": 2.1381818181818182e-05,
|
| 355 |
+
"loss": 2.0026,
|
| 356 |
+
"step": 110
|
| 357 |
},
|
| 358 |
{
|
| 359 |
+
"epoch": 1.09,
|
| 360 |
+
"learning_rate": 2.1163636363636364e-05,
|
| 361 |
+
"loss": 1.9867,
|
| 362 |
+
"step": 112
|
| 363 |
},
|
| 364 |
{
|
| 365 |
+
"epoch": 1.11,
|
| 366 |
+
"learning_rate": 2.0945454545454546e-05,
|
| 367 |
+
"loss": 2.0086,
|
| 368 |
+
"step": 114
|
| 369 |
},
|
| 370 |
{
|
| 371 |
+
"epoch": 1.13,
|
| 372 |
+
"learning_rate": 2.0727272727272728e-05,
|
| 373 |
+
"loss": 1.9611,
|
| 374 |
+
"step": 116
|
| 375 |
},
|
| 376 |
{
|
| 377 |
+
"epoch": 1.15,
|
| 378 |
+
"learning_rate": 2.0509090909090907e-05,
|
| 379 |
+
"loss": 1.9681,
|
| 380 |
+
"step": 118
|
| 381 |
},
|
| 382 |
{
|
| 383 |
+
"epoch": 1.17,
|
| 384 |
+
"learning_rate": 2.0290909090909092e-05,
|
| 385 |
+
"loss": 1.9623,
|
| 386 |
+
"step": 120
|
| 387 |
},
|
| 388 |
{
|
| 389 |
+
"epoch": 1.19,
|
| 390 |
+
"learning_rate": 2.0072727272727274e-05,
|
| 391 |
+
"loss": 1.9776,
|
| 392 |
+
"step": 122
|
| 393 |
},
|
| 394 |
{
|
| 395 |
+
"epoch": 1.21,
|
| 396 |
+
"learning_rate": 1.9854545454545456e-05,
|
| 397 |
+
"loss": 1.999,
|
| 398 |
+
"step": 124
|
| 399 |
},
|
| 400 |
{
|
| 401 |
+
"epoch": 1.23,
|
| 402 |
+
"learning_rate": 1.963636363636364e-05,
|
| 403 |
+
"loss": 1.9827,
|
| 404 |
+
"step": 126
|
| 405 |
},
|
| 406 |
{
|
| 407 |
+
"epoch": 1.25,
|
| 408 |
+
"learning_rate": 1.9418181818181817e-05,
|
| 409 |
+
"loss": 1.9945,
|
| 410 |
+
"step": 128
|
| 411 |
},
|
| 412 |
{
|
| 413 |
+
"epoch": 1.27,
|
| 414 |
+
"learning_rate": 1.9200000000000003e-05,
|
| 415 |
+
"loss": 1.9692,
|
| 416 |
+
"step": 130
|
| 417 |
},
|
| 418 |
{
|
| 419 |
+
"epoch": 1.29,
|
| 420 |
+
"learning_rate": 1.898181818181818e-05,
|
| 421 |
+
"loss": 1.9519,
|
| 422 |
+
"step": 132
|
| 423 |
},
|
| 424 |
{
|
| 425 |
+
"epoch": 1.31,
|
| 426 |
+
"learning_rate": 1.8763636363636367e-05,
|
| 427 |
+
"loss": 1.9432,
|
| 428 |
+
"step": 134
|
| 429 |
},
|
| 430 |
{
|
| 431 |
+
"epoch": 1.33,
|
| 432 |
+
"learning_rate": 1.8545454545454545e-05,
|
| 433 |
+
"loss": 1.9454,
|
| 434 |
+
"step": 136
|
| 435 |
},
|
| 436 |
{
|
| 437 |
+
"epoch": 1.35,
|
| 438 |
+
"learning_rate": 1.832727272727273e-05,
|
| 439 |
+
"loss": 1.9759,
|
| 440 |
+
"step": 138
|
|
|
|
|
|
|
|
|
|
| 441 |
},
|
| 442 |
{
|
| 443 |
+
"epoch": 1.37,
|
| 444 |
+
"learning_rate": 1.810909090909091e-05,
|
| 445 |
+
"loss": 1.9672,
|
| 446 |
+
"step": 140
|
| 447 |
},
|
| 448 |
{
|
| 449 |
+
"epoch": 1.39,
|
| 450 |
+
"learning_rate": 1.789090909090909e-05,
|
| 451 |
+
"loss": 1.9527,
|
| 452 |
+
"step": 142
|
| 453 |
},
|
| 454 |
{
|
| 455 |
+
"epoch": 1.41,
|
| 456 |
+
"learning_rate": 1.7672727272727274e-05,
|
| 457 |
+
"loss": 1.9465,
|
| 458 |
+
"step": 144
|
| 459 |
},
|
| 460 |
{
|
| 461 |
+
"epoch": 1.43,
|
| 462 |
+
"learning_rate": 1.7454545454545456e-05,
|
| 463 |
+
"loss": 1.9948,
|
| 464 |
+
"step": 146
|
| 465 |
},
|
| 466 |
{
|
| 467 |
+
"epoch": 1.44,
|
| 468 |
+
"learning_rate": 1.7236363636363638e-05,
|
| 469 |
+
"loss": 1.9685,
|
| 470 |
+
"step": 148
|
| 471 |
},
|
| 472 |
{
|
| 473 |
+
"epoch": 1.46,
|
| 474 |
+
"learning_rate": 1.701818181818182e-05,
|
| 475 |
+
"loss": 1.9518,
|
| 476 |
+
"step": 150
|
| 477 |
},
|
| 478 |
{
|
| 479 |
+
"epoch": 1.48,
|
| 480 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 481 |
+
"loss": 1.9624,
|
| 482 |
+
"step": 152
|
| 483 |
},
|
| 484 |
{
|
| 485 |
+
"epoch": 1.49,
|
| 486 |
+
"eval_accuracy": 0.6862566888822462,
|
| 487 |
+
"eval_loss": 2.0935797691345215,
|
| 488 |
+
"eval_runtime": 510.8965,
|
| 489 |
+
"eval_samples_per_second": 3.515,
|
| 490 |
+
"eval_steps_per_second": 3.515,
|
| 491 |
+
"step": 153
|
| 492 |
},
|
| 493 |
{
|
| 494 |
+
"epoch": 1.5,
|
| 495 |
+
"learning_rate": 1.6581818181818184e-05,
|
| 496 |
+
"loss": 1.9587,
|
| 497 |
+
"step": 154
|
| 498 |
},
|
| 499 |
{
|
| 500 |
+
"epoch": 1.52,
|
| 501 |
+
"learning_rate": 1.6363636363636363e-05,
|
| 502 |
+
"loss": 1.9752,
|
| 503 |
+
"step": 156
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"epoch": 1.54,
|
| 507 |
+
"learning_rate": 1.6145454545454548e-05,
|
| 508 |
+
"loss": 1.9559,
|
| 509 |
+
"step": 158
|
| 510 |
},
|
| 511 |
{
|
| 512 |
+
"epoch": 1.56,
|
| 513 |
+
"learning_rate": 1.5927272727272727e-05,
|
| 514 |
+
"loss": 1.9707,
|
| 515 |
+
"step": 160
|
| 516 |
},
|
| 517 |
{
|
| 518 |
+
"epoch": 1.58,
|
| 519 |
+
"learning_rate": 1.5709090909090912e-05,
|
| 520 |
+
"loss": 1.9844,
|
| 521 |
+
"step": 162
|
| 522 |
},
|
| 523 |
{
|
| 524 |
+
"epoch": 1.6,
|
| 525 |
+
"learning_rate": 1.549090909090909e-05,
|
| 526 |
+
"loss": 1.9884,
|
| 527 |
+
"step": 164
|
| 528 |
},
|
| 529 |
{
|
| 530 |
+
"epoch": 1.62,
|
| 531 |
+
"learning_rate": 1.5272727272727273e-05,
|
| 532 |
+
"loss": 1.9799,
|
| 533 |
+
"step": 166
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"epoch": 1.64,
|
| 537 |
+
"learning_rate": 1.5054545454545455e-05,
|
| 538 |
+
"loss": 1.9582,
|
| 539 |
+
"step": 168
|
| 540 |
},
|
| 541 |
{
|
| 542 |
+
"epoch": 1.66,
|
| 543 |
+
"learning_rate": 1.4836363636363637e-05,
|
| 544 |
+
"loss": 1.9677,
|
| 545 |
+
"step": 170
|
| 546 |
},
|
| 547 |
{
|
| 548 |
+
"epoch": 1.68,
|
| 549 |
+
"learning_rate": 1.4618181818181817e-05,
|
| 550 |
+
"loss": 1.9804,
|
| 551 |
+
"step": 172
|
|
|
|
|
|
|
|
|
|
| 552 |
},
|
| 553 |
{
|
| 554 |
+
"epoch": 1.7,
|
| 555 |
+
"learning_rate": 1.44e-05,
|
| 556 |
+
"loss": 1.9816,
|
| 557 |
+
"step": 174
|
| 558 |
},
|
| 559 |
{
|
| 560 |
+
"epoch": 1.72,
|
| 561 |
+
"learning_rate": 1.4181818181818181e-05,
|
| 562 |
+
"loss": 1.969,
|
| 563 |
+
"step": 176
|
| 564 |
},
|
| 565 |
{
|
| 566 |
+
"epoch": 1.74,
|
| 567 |
+
"learning_rate": 1.3963636363636363e-05,
|
| 568 |
+
"loss": 1.9602,
|
| 569 |
+
"step": 178
|
| 570 |
},
|
| 571 |
{
|
| 572 |
+
"epoch": 1.76,
|
| 573 |
+
"learning_rate": 1.3745454545454546e-05,
|
| 574 |
+
"loss": 1.9926,
|
| 575 |
+
"step": 180
|
| 576 |
},
|
| 577 |
{
|
| 578 |
+
"epoch": 1.78,
|
| 579 |
+
"learning_rate": 1.3527272727272728e-05,
|
| 580 |
+
"loss": 1.9378,
|
| 581 |
+
"step": 182
|
| 582 |
},
|
| 583 |
{
|
| 584 |
+
"epoch": 1.8,
|
| 585 |
+
"learning_rate": 1.330909090909091e-05,
|
| 586 |
+
"loss": 1.9567,
|
| 587 |
+
"step": 184
|
| 588 |
},
|
| 589 |
{
|
| 590 |
+
"epoch": 1.82,
|
| 591 |
+
"learning_rate": 1.309090909090909e-05,
|
| 592 |
+
"loss": 1.9849,
|
| 593 |
+
"step": 186
|
| 594 |
},
|
| 595 |
{
|
| 596 |
+
"epoch": 1.84,
|
| 597 |
+
"learning_rate": 1.2872727272727272e-05,
|
| 598 |
+
"loss": 1.9622,
|
| 599 |
+
"step": 188
|
| 600 |
},
|
| 601 |
{
|
| 602 |
+
"epoch": 1.85,
|
| 603 |
+
"learning_rate": 1.2654545454545454e-05,
|
| 604 |
+
"loss": 1.9447,
|
| 605 |
+
"step": 190
|
| 606 |
},
|
| 607 |
{
|
| 608 |
+
"epoch": 1.87,
|
| 609 |
+
"learning_rate": 1.2436363636363636e-05,
|
| 610 |
+
"loss": 1.958,
|
| 611 |
+
"step": 192
|
| 612 |
},
|
| 613 |
{
|
| 614 |
+
"epoch": 1.89,
|
| 615 |
+
"learning_rate": 1.2218181818181818e-05,
|
| 616 |
+
"loss": 1.9729,
|
| 617 |
+
"step": 194
|
| 618 |
},
|
| 619 |
{
|
| 620 |
+
"epoch": 1.91,
|
| 621 |
+
"learning_rate": 1.2e-05,
|
| 622 |
+
"loss": 1.9402,
|
| 623 |
+
"step": 196
|
| 624 |
},
|
| 625 |
{
|
| 626 |
+
"epoch": 1.93,
|
| 627 |
+
"learning_rate": 1.178181818181818e-05,
|
| 628 |
+
"loss": 1.977,
|
| 629 |
+
"step": 198
|
| 630 |
},
|
| 631 |
{
|
| 632 |
+
"epoch": 1.95,
|
| 633 |
+
"learning_rate": 1.1563636363636363e-05,
|
| 634 |
+
"loss": 1.9561,
|
| 635 |
+
"step": 200
|
| 636 |
},
|
| 637 |
{
|
| 638 |
+
"epoch": 1.97,
|
| 639 |
+
"learning_rate": 1.1345454545454545e-05,
|
| 640 |
+
"loss": 1.9576,
|
| 641 |
+
"step": 202
|
| 642 |
},
|
| 643 |
{
|
| 644 |
+
"epoch": 1.99,
|
| 645 |
+
"learning_rate": 1.1127272727272727e-05,
|
| 646 |
+
"loss": 1.9476,
|
| 647 |
+
"step": 204
|
| 648 |
},
|
| 649 |
{
|
| 650 |
+
"epoch": 1.99,
|
| 651 |
+
"eval_accuracy": 0.6870071488415761,
|
| 652 |
+
"eval_loss": 2.0888001918792725,
|
| 653 |
+
"eval_runtime": 510.9696,
|
| 654 |
+
"eval_samples_per_second": 3.515,
|
| 655 |
+
"eval_steps_per_second": 3.515,
|
| 656 |
+
"step": 204
|
| 657 |
},
|
| 658 |
{
|
| 659 |
+
"epoch": 2.01,
|
| 660 |
+
"learning_rate": 1.0909090909090909e-05,
|
| 661 |
+
"loss": 1.9169,
|
| 662 |
+
"step": 206
|
| 663 |
},
|
| 664 |
{
|
| 665 |
+
"epoch": 2.03,
|
| 666 |
+
"learning_rate": 1.0690909090909091e-05,
|
| 667 |
+
"loss": 1.8484,
|
| 668 |
+
"step": 208
|
| 669 |
},
|
| 670 |
{
|
| 671 |
+
"epoch": 2.05,
|
| 672 |
+
"learning_rate": 1.0472727272727273e-05,
|
| 673 |
+
"loss": 1.8525,
|
| 674 |
+
"step": 210
|
| 675 |
},
|
| 676 |
{
|
| 677 |
+
"epoch": 2.07,
|
| 678 |
+
"learning_rate": 1.0254545454545453e-05,
|
| 679 |
+
"loss": 1.8692,
|
| 680 |
+
"step": 212
|
| 681 |
},
|
| 682 |
{
|
| 683 |
+
"epoch": 2.09,
|
| 684 |
+
"learning_rate": 1.0036363636363637e-05,
|
| 685 |
+
"loss": 1.8738,
|
| 686 |
+
"step": 214
|
| 687 |
},
|
| 688 |
{
|
| 689 |
+
"epoch": 2.11,
|
| 690 |
+
"learning_rate": 9.81818181818182e-06,
|
| 691 |
+
"loss": 1.8889,
|
| 692 |
+
"step": 216
|
| 693 |
},
|
| 694 |
{
|
| 695 |
+
"epoch": 2.13,
|
| 696 |
+
"learning_rate": 9.600000000000001e-06,
|
| 697 |
+
"loss": 1.8661,
|
| 698 |
+
"step": 218
|
| 699 |
},
|
| 700 |
{
|
| 701 |
+
"epoch": 2.15,
|
| 702 |
+
"learning_rate": 9.381818181818183e-06,
|
| 703 |
+
"loss": 1.8699,
|
| 704 |
+
"step": 220
|
| 705 |
},
|
| 706 |
{
|
| 707 |
+
"epoch": 2.17,
|
| 708 |
+
"learning_rate": 9.163636363636365e-06,
|
| 709 |
+
"loss": 1.8902,
|
| 710 |
+
"step": 222
|
| 711 |
},
|
| 712 |
{
|
| 713 |
+
"epoch": 2.19,
|
| 714 |
+
"learning_rate": 8.945454545454546e-06,
|
| 715 |
+
"loss": 1.8766,
|
| 716 |
+
"step": 224
|
| 717 |
},
|
| 718 |
{
|
| 719 |
+
"epoch": 2.21,
|
| 720 |
+
"learning_rate": 8.727272727272728e-06,
|
| 721 |
+
"loss": 1.8634,
|
| 722 |
+
"step": 226
|
| 723 |
},
|
| 724 |
{
|
| 725 |
+
"epoch": 2.23,
|
| 726 |
+
"learning_rate": 8.50909090909091e-06,
|
| 727 |
+
"loss": 1.8549,
|
| 728 |
+
"step": 228
|
| 729 |
},
|
| 730 |
{
|
| 731 |
+
"epoch": 2.25,
|
| 732 |
+
"learning_rate": 8.290909090909092e-06,
|
| 733 |
+
"loss": 1.8592,
|
| 734 |
+
"step": 230
|
| 735 |
},
|
| 736 |
{
|
| 737 |
+
"epoch": 2.26,
|
| 738 |
+
"learning_rate": 8.072727272727274e-06,
|
| 739 |
+
"loss": 1.8695,
|
| 740 |
+
"step": 232
|
| 741 |
},
|
| 742 |
{
|
| 743 |
+
"epoch": 2.28,
|
| 744 |
+
"learning_rate": 7.854545454545456e-06,
|
| 745 |
+
"loss": 1.8753,
|
| 746 |
+
"step": 234
|
| 747 |
},
|
| 748 |
{
|
| 749 |
+
"epoch": 2.3,
|
| 750 |
+
"learning_rate": 7.636363636363636e-06,
|
| 751 |
+
"loss": 1.8599,
|
| 752 |
+
"step": 236
|
| 753 |
},
|
| 754 |
{
|
| 755 |
+
"epoch": 2.32,
|
| 756 |
+
"learning_rate": 7.4181818181818185e-06,
|
| 757 |
+
"loss": 1.8847,
|
| 758 |
+
"step": 238
|
|
|
|
|
|
|
|
|
|
| 759 |
},
|
| 760 |
{
|
| 761 |
+
"epoch": 2.34,
|
| 762 |
+
"learning_rate": 7.2e-06,
|
| 763 |
+
"loss": 1.8712,
|
| 764 |
+
"step": 240
|
| 765 |
},
|
| 766 |
{
|
| 767 |
+
"epoch": 2.36,
|
| 768 |
+
"learning_rate": 6.981818181818182e-06,
|
| 769 |
+
"loss": 1.8673,
|
| 770 |
+
"step": 242
|
| 771 |
},
|
| 772 |
{
|
| 773 |
+
"epoch": 2.38,
|
| 774 |
+
"learning_rate": 6.763636363636364e-06,
|
| 775 |
+
"loss": 1.8617,
|
| 776 |
+
"step": 244
|
| 777 |
},
|
| 778 |
{
|
| 779 |
+
"epoch": 2.4,
|
| 780 |
+
"learning_rate": 6.545454545454545e-06,
|
| 781 |
+
"loss": 1.8297,
|
| 782 |
+
"step": 246
|
| 783 |
},
|
| 784 |
{
|
| 785 |
+
"epoch": 2.42,
|
| 786 |
+
"learning_rate": 6.327272727272727e-06,
|
| 787 |
+
"loss": 1.8624,
|
| 788 |
+
"step": 248
|
| 789 |
},
|
| 790 |
{
|
| 791 |
+
"epoch": 2.44,
|
| 792 |
+
"learning_rate": 6.109090909090909e-06,
|
| 793 |
+
"loss": 1.8753,
|
| 794 |
+
"step": 250
|
| 795 |
},
|
| 796 |
{
|
| 797 |
+
"epoch": 2.46,
|
| 798 |
+
"learning_rate": 5.89090909090909e-06,
|
| 799 |
+
"loss": 1.8789,
|
| 800 |
+
"step": 252
|
| 801 |
},
|
| 802 |
{
|
| 803 |
+
"epoch": 2.48,
|
| 804 |
+
"learning_rate": 5.672727272727272e-06,
|
| 805 |
+
"loss": 1.888,
|
| 806 |
+
"step": 254
|
| 807 |
},
|
| 808 |
{
|
| 809 |
+
"epoch": 2.49,
|
| 810 |
+
"eval_accuracy": 0.6863803621574514,
|
| 811 |
+
"eval_loss": 2.0992937088012695,
|
| 812 |
+
"eval_runtime": 511.1737,
|
| 813 |
+
"eval_samples_per_second": 3.513,
|
| 814 |
+
"eval_steps_per_second": 3.513,
|
| 815 |
+
"step": 255
|
| 816 |
},
|
| 817 |
{
|
| 818 |
+
"epoch": 2.5,
|
| 819 |
+
"learning_rate": 5.4545454545454545e-06,
|
| 820 |
+
"loss": 1.8462,
|
| 821 |
+
"step": 256
|
| 822 |
},
|
| 823 |
{
|
| 824 |
+
"epoch": 2.52,
|
| 825 |
+
"learning_rate": 5.2363636363636365e-06,
|
| 826 |
+
"loss": 1.8716,
|
| 827 |
+
"step": 258
|
| 828 |
},
|
| 829 |
{
|
| 830 |
+
"epoch": 2.54,
|
| 831 |
+
"learning_rate": 5.0181818181818186e-06,
|
| 832 |
+
"loss": 1.8823,
|
| 833 |
+
"step": 260
|
| 834 |
},
|
| 835 |
{
|
| 836 |
+
"epoch": 2.56,
|
| 837 |
+
"learning_rate": 4.800000000000001e-06,
|
| 838 |
+
"loss": 1.8718,
|
| 839 |
+
"step": 262
|
| 840 |
},
|
| 841 |
{
|
| 842 |
+
"epoch": 2.58,
|
| 843 |
+
"learning_rate": 4.581818181818183e-06,
|
| 844 |
+
"loss": 1.8745,
|
| 845 |
+
"step": 264
|
| 846 |
},
|
| 847 |
{
|
| 848 |
+
"epoch": 2.6,
|
| 849 |
+
"learning_rate": 4.363636363636364e-06,
|
| 850 |
+
"loss": 1.908,
|
| 851 |
+
"step": 266
|
| 852 |
},
|
| 853 |
{
|
| 854 |
+
"epoch": 2.62,
|
| 855 |
+
"learning_rate": 4.145454545454546e-06,
|
| 856 |
+
"loss": 1.8412,
|
| 857 |
+
"step": 268
|
| 858 |
},
|
| 859 |
{
|
| 860 |
+
"epoch": 2.64,
|
| 861 |
+
"learning_rate": 3.927272727272728e-06,
|
| 862 |
+
"loss": 1.8462,
|
| 863 |
+
"step": 270
|
| 864 |
},
|
| 865 |
{
|
| 866 |
+
"epoch": 2.66,
|
| 867 |
+
"learning_rate": 3.7090909090909092e-06,
|
| 868 |
+
"loss": 1.8683,
|
| 869 |
+
"step": 272
|
|
|
|
|
|
|
|
|
|
| 870 |
},
|
| 871 |
{
|
| 872 |
+
"epoch": 2.68,
|
| 873 |
+
"learning_rate": 3.490909090909091e-06,
|
| 874 |
+
"loss": 1.872,
|
| 875 |
+
"step": 274
|
| 876 |
},
|
| 877 |
{
|
| 878 |
+
"epoch": 2.69,
|
| 879 |
+
"learning_rate": 3.2727272727272725e-06,
|
| 880 |
+
"loss": 1.852,
|
| 881 |
+
"step": 276
|
| 882 |
},
|
| 883 |
{
|
| 884 |
+
"epoch": 2.71,
|
| 885 |
+
"learning_rate": 3.0545454545454546e-06,
|
| 886 |
+
"loss": 1.8619,
|
| 887 |
+
"step": 278
|
| 888 |
},
|
| 889 |
{
|
| 890 |
+
"epoch": 2.73,
|
| 891 |
+
"learning_rate": 2.836363636363636e-06,
|
| 892 |
+
"loss": 1.8542,
|
| 893 |
+
"step": 280
|
| 894 |
},
|
| 895 |
{
|
| 896 |
+
"epoch": 2.75,
|
| 897 |
+
"learning_rate": 2.6181818181818183e-06,
|
| 898 |
+
"loss": 1.8953,
|
| 899 |
+
"step": 282
|
| 900 |
},
|
| 901 |
{
|
| 902 |
+
"epoch": 2.77,
|
| 903 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 904 |
+
"loss": 1.8341,
|
| 905 |
+
"step": 284
|
| 906 |
},
|
| 907 |
{
|
| 908 |
+
"epoch": 2.79,
|
| 909 |
+
"learning_rate": 2.181818181818182e-06,
|
| 910 |
+
"loss": 1.8484,
|
| 911 |
+
"step": 286
|
| 912 |
},
|
| 913 |
{
|
| 914 |
+
"epoch": 2.81,
|
| 915 |
+
"learning_rate": 1.963636363636364e-06,
|
| 916 |
+
"loss": 1.8473,
|
| 917 |
+
"step": 288
|
| 918 |
},
|
| 919 |
{
|
| 920 |
+
"epoch": 2.83,
|
| 921 |
+
"learning_rate": 1.7454545454545454e-06,
|
| 922 |
+
"loss": 1.8827,
|
| 923 |
+
"step": 290
|
| 924 |
},
|
| 925 |
{
|
| 926 |
+
"epoch": 2.85,
|
| 927 |
+
"learning_rate": 1.5272727272727273e-06,
|
| 928 |
+
"loss": 1.8572,
|
| 929 |
+
"step": 292
|
| 930 |
},
|
| 931 |
{
|
| 932 |
+
"epoch": 2.87,
|
| 933 |
+
"learning_rate": 1.3090909090909091e-06,
|
| 934 |
+
"loss": 1.8721,
|
| 935 |
+
"step": 294
|
| 936 |
},
|
| 937 |
{
|
| 938 |
+
"epoch": 2.89,
|
| 939 |
+
"learning_rate": 1.090909090909091e-06,
|
| 940 |
+
"loss": 1.8363,
|
| 941 |
+
"step": 296
|
| 942 |
},
|
| 943 |
{
|
| 944 |
+
"epoch": 2.91,
|
| 945 |
+
"learning_rate": 8.727272727272727e-07,
|
| 946 |
+
"loss": 1.8762,
|
| 947 |
+
"step": 298
|
| 948 |
},
|
| 949 |
{
|
| 950 |
+
"epoch": 2.93,
|
| 951 |
+
"learning_rate": 6.545454545454546e-07,
|
| 952 |
+
"loss": 1.8749,
|
| 953 |
+
"step": 300
|
| 954 |
},
|
| 955 |
{
|
| 956 |
+
"epoch": 2.95,
|
| 957 |
+
"learning_rate": 4.3636363636363636e-07,
|
| 958 |
+
"loss": 1.8522,
|
| 959 |
+
"step": 302
|
| 960 |
},
|
| 961 |
{
|
| 962 |
+
"epoch": 2.97,
|
| 963 |
+
"learning_rate": 2.1818181818181818e-07,
|
| 964 |
+
"loss": 1.8716,
|
| 965 |
+
"step": 304
|
| 966 |
},
|
| 967 |
{
|
| 968 |
+
"epoch": 2.99,
|
| 969 |
+
"learning_rate": 0.0,
|
| 970 |
+
"loss": 1.8687,
|
| 971 |
+
"step": 306
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 2.99,
|
| 975 |
+
"eval_accuracy": 0.6864508822377811,
|
| 976 |
+
"eval_loss": 2.0993993282318115,
|
| 977 |
+
"eval_runtime": 511.0772,
|
| 978 |
+
"eval_samples_per_second": 3.514,
|
| 979 |
+
"eval_steps_per_second": 3.514,
|
| 980 |
+
"step": 306
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 2.99,
|
| 984 |
+
"step": 306,
|
| 985 |
+
"total_flos": 4.2741192183760896e+17,
|
| 986 |
+
"train_loss": 2.010221361334807,
|
| 987 |
+
"train_runtime": 46544.0603,
|
| 988 |
+
"train_samples_per_second": 0.845,
|
| 989 |
+
"train_steps_per_second": 0.007
|
| 990 |
}
|
| 991 |
],
|
| 992 |
+
"max_steps": 306,
|
| 993 |
+
"num_train_epochs": 3,
|
| 994 |
+
"total_flos": 4.2741192183760896e+17,
|
| 995 |
"trial_name": null,
|
| 996 |
"trial_params": null
|
| 997 |
}
|
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:a71ef2368d7982b99f77d7631603c4ef031cdc2d890bf705f3324ac76a0bcffc
|
| 3 |
size 3579
|