Commit
•
e4ef258
1
Parent(s):
a9e18da
Model save
Browse files- README.md +78 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +436 -0
README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3.1
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Llama-31-8B_task-2_120-samples_config-2_auto
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# Llama-31-8B_task-2_120-samples_config-2_auto
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.0912
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0001
|
| 41 |
+
- train_batch_size: 1
|
| 42 |
+
- eval_batch_size: 1
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- gradient_accumulation_steps: 16
|
| 46 |
+
- total_train_batch_size: 16
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
+
- num_epochs: 50
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-------:|:----:|:---------------:|
|
| 56 |
+
| 0.9245 | 0.9091 | 5 | 0.8871 |
|
| 57 |
+
| 0.8834 | 2.0 | 11 | 0.7984 |
|
| 58 |
+
| 0.7224 | 2.9091 | 16 | 0.7230 |
|
| 59 |
+
| 0.6496 | 4.0 | 22 | 0.6428 |
|
| 60 |
+
| 0.5918 | 4.9091 | 27 | 0.6049 |
|
| 61 |
+
| 0.5484 | 6.0 | 33 | 0.5821 |
|
| 62 |
+
| 0.472 | 6.9091 | 38 | 0.5819 |
|
| 63 |
+
| 0.4328 | 8.0 | 44 | 0.5885 |
|
| 64 |
+
| 0.3688 | 8.9091 | 49 | 0.6169 |
|
| 65 |
+
| 0.2923 | 10.0 | 55 | 0.6638 |
|
| 66 |
+
| 0.2063 | 10.9091 | 60 | 0.7679 |
|
| 67 |
+
| 0.1206 | 12.0 | 66 | 0.8585 |
|
| 68 |
+
| 0.0643 | 12.9091 | 71 | 1.0241 |
|
| 69 |
+
| 0.0373 | 14.0 | 77 | 1.0912 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- PEFT 0.12.0
|
| 75 |
+
- Transformers 4.44.0
|
| 76 |
+
- Pytorch 2.1.2+cu121
|
| 77 |
+
- Datasets 2.20.0
|
| 78 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 167832240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c64e1ac68ac08ec206fa484b14abb579448e10801e0e822bf97e1986d1e0452e
|
| 3 |
size 167832240
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"total_flos": 1.219390286677934e+17,
|
| 4 |
+
"train_loss": 0.4532604143023491,
|
| 5 |
+
"train_runtime": 4348.0981,
|
| 6 |
+
"train_samples": 88,
|
| 7 |
+
"train_samples_per_second": 1.012,
|
| 8 |
+
"train_steps_per_second": 0.057
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"total_flos": 1.219390286677934e+17,
|
| 4 |
+
"train_loss": 0.4532604143023491,
|
| 5 |
+
"train_runtime": 4348.0981,
|
| 6 |
+
"train_samples": 88,
|
| 7 |
+
"train_samples_per_second": 1.012,
|
| 8 |
+
"train_steps_per_second": 0.057
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,436 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.58187335729599,
|
| 3 |
+
"best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-2_auto/checkpoint-38",
|
| 4 |
+
"epoch": 14.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 77,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.18181818181818182,
|
| 13 |
+
"grad_norm": 0.4114352762699127,
|
| 14 |
+
"learning_rate": 4.000000000000001e-06,
|
| 15 |
+
"loss": 0.9261,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.36363636363636365,
|
| 20 |
+
"grad_norm": 0.36285364627838135,
|
| 21 |
+
"learning_rate": 8.000000000000001e-06,
|
| 22 |
+
"loss": 0.8467,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7272727272727273,
|
| 27 |
+
"grad_norm": 0.39619794487953186,
|
| 28 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 29 |
+
"loss": 0.9245,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9090909090909091,
|
| 34 |
+
"eval_loss": 0.887127697467804,
|
| 35 |
+
"eval_runtime": 25.0122,
|
| 36 |
+
"eval_samples_per_second": 0.96,
|
| 37 |
+
"eval_steps_per_second": 0.96,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0909090909090908,
|
| 42 |
+
"grad_norm": 0.39810460805892944,
|
| 43 |
+
"learning_rate": 2.4e-05,
|
| 44 |
+
"loss": 0.887,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.4545454545454546,
|
| 49 |
+
"grad_norm": 0.39191126823425293,
|
| 50 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 51 |
+
"loss": 0.8678,
|
| 52 |
+
"step": 8
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.8181818181818183,
|
| 56 |
+
"grad_norm": 0.3597930073738098,
|
| 57 |
+
"learning_rate": 4e-05,
|
| 58 |
+
"loss": 0.8834,
|
| 59 |
+
"step": 10
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.0,
|
| 63 |
+
"eval_loss": 0.7983560562133789,
|
| 64 |
+
"eval_runtime": 25.0057,
|
| 65 |
+
"eval_samples_per_second": 0.96,
|
| 66 |
+
"eval_steps_per_second": 0.96,
|
| 67 |
+
"step": 11
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.1818181818181817,
|
| 71 |
+
"grad_norm": 0.30901816487312317,
|
| 72 |
+
"learning_rate": 4.8e-05,
|
| 73 |
+
"loss": 0.752,
|
| 74 |
+
"step": 12
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.5454545454545454,
|
| 78 |
+
"grad_norm": 0.278992235660553,
|
| 79 |
+
"learning_rate": 5.6000000000000006e-05,
|
| 80 |
+
"loss": 0.7998,
|
| 81 |
+
"step": 14
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.909090909090909,
|
| 85 |
+
"grad_norm": 0.3216788172721863,
|
| 86 |
+
"learning_rate": 6.400000000000001e-05,
|
| 87 |
+
"loss": 0.7224,
|
| 88 |
+
"step": 16
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.909090909090909,
|
| 92 |
+
"eval_loss": 0.7229611277580261,
|
| 93 |
+
"eval_runtime": 25.0099,
|
| 94 |
+
"eval_samples_per_second": 0.96,
|
| 95 |
+
"eval_steps_per_second": 0.96,
|
| 96 |
+
"step": 16
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.2727272727272725,
|
| 100 |
+
"grad_norm": 0.34014081954956055,
|
| 101 |
+
"learning_rate": 7.2e-05,
|
| 102 |
+
"loss": 0.6977,
|
| 103 |
+
"step": 18
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 3.6363636363636362,
|
| 107 |
+
"grad_norm": 0.2702198028564453,
|
| 108 |
+
"learning_rate": 8e-05,
|
| 109 |
+
"loss": 0.6652,
|
| 110 |
+
"step": 20
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.0,
|
| 114 |
+
"grad_norm": 0.2623775601387024,
|
| 115 |
+
"learning_rate": 8.800000000000001e-05,
|
| 116 |
+
"loss": 0.6496,
|
| 117 |
+
"step": 22
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 4.0,
|
| 121 |
+
"eval_loss": 0.6428138613700867,
|
| 122 |
+
"eval_runtime": 25.008,
|
| 123 |
+
"eval_samples_per_second": 0.96,
|
| 124 |
+
"eval_steps_per_second": 0.96,
|
| 125 |
+
"step": 22
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.363636363636363,
|
| 129 |
+
"grad_norm": 0.27580147981643677,
|
| 130 |
+
"learning_rate": 9.6e-05,
|
| 131 |
+
"loss": 0.6274,
|
| 132 |
+
"step": 24
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.7272727272727275,
|
| 136 |
+
"grad_norm": 0.3490183651447296,
|
| 137 |
+
"learning_rate": 9.999512620046522e-05,
|
| 138 |
+
"loss": 0.5918,
|
| 139 |
+
"step": 26
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 4.909090909090909,
|
| 143 |
+
"eval_loss": 0.6048721671104431,
|
| 144 |
+
"eval_runtime": 25.0083,
|
| 145 |
+
"eval_samples_per_second": 0.96,
|
| 146 |
+
"eval_steps_per_second": 0.96,
|
| 147 |
+
"step": 27
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.090909090909091,
|
| 151 |
+
"grad_norm": 0.28270548582077026,
|
| 152 |
+
"learning_rate": 9.995614150494293e-05,
|
| 153 |
+
"loss": 0.564,
|
| 154 |
+
"step": 28
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.454545454545454,
|
| 158 |
+
"grad_norm": 0.294200599193573,
|
| 159 |
+
"learning_rate": 9.987820251299122e-05,
|
| 160 |
+
"loss": 0.5366,
|
| 161 |
+
"step": 30
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.818181818181818,
|
| 165 |
+
"grad_norm": 0.29604193568229675,
|
| 166 |
+
"learning_rate": 9.976136999909156e-05,
|
| 167 |
+
"loss": 0.5484,
|
| 168 |
+
"step": 32
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.0,
|
| 172 |
+
"eval_loss": 0.5820669531822205,
|
| 173 |
+
"eval_runtime": 25.0148,
|
| 174 |
+
"eval_samples_per_second": 0.959,
|
| 175 |
+
"eval_steps_per_second": 0.959,
|
| 176 |
+
"step": 33
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.181818181818182,
|
| 180 |
+
"grad_norm": 0.25547143816947937,
|
| 181 |
+
"learning_rate": 9.96057350657239e-05,
|
| 182 |
+
"loss": 0.5068,
|
| 183 |
+
"step": 34
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.545454545454545,
|
| 187 |
+
"grad_norm": 0.2587781846523285,
|
| 188 |
+
"learning_rate": 9.941141907232765e-05,
|
| 189 |
+
"loss": 0.5096,
|
| 190 |
+
"step": 36
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.909090909090909,
|
| 194 |
+
"grad_norm": 0.30631354451179504,
|
| 195 |
+
"learning_rate": 9.917857354066931e-05,
|
| 196 |
+
"loss": 0.472,
|
| 197 |
+
"step": 38
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 6.909090909090909,
|
| 201 |
+
"eval_loss": 0.58187335729599,
|
| 202 |
+
"eval_runtime": 25.0075,
|
| 203 |
+
"eval_samples_per_second": 0.96,
|
| 204 |
+
"eval_steps_per_second": 0.96,
|
| 205 |
+
"step": 38
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.2727272727272725,
|
| 209 |
+
"grad_norm": 0.2765539586544037,
|
| 210 |
+
"learning_rate": 9.890738003669029e-05,
|
| 211 |
+
"loss": 0.4694,
|
| 212 |
+
"step": 40
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.636363636363637,
|
| 216 |
+
"grad_norm": 0.307478129863739,
|
| 217 |
+
"learning_rate": 9.859805002892732e-05,
|
| 218 |
+
"loss": 0.4066,
|
| 219 |
+
"step": 42
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 8.0,
|
| 223 |
+
"grad_norm": 0.3184185326099396,
|
| 224 |
+
"learning_rate": 9.825082472361557e-05,
|
| 225 |
+
"loss": 0.4328,
|
| 226 |
+
"step": 44
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_loss": 0.5885212421417236,
|
| 231 |
+
"eval_runtime": 25.006,
|
| 232 |
+
"eval_samples_per_second": 0.96,
|
| 233 |
+
"eval_steps_per_second": 0.96,
|
| 234 |
+
"step": 44
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.363636363636363,
|
| 238 |
+
"grad_norm": 0.3068244159221649,
|
| 239 |
+
"learning_rate": 9.786597487660337e-05,
|
| 240 |
+
"loss": 0.3683,
|
| 241 |
+
"step": 46
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 8.727272727272727,
|
| 245 |
+
"grad_norm": 0.3444615602493286,
|
| 246 |
+
"learning_rate": 9.744380058222483e-05,
|
| 247 |
+
"loss": 0.3688,
|
| 248 |
+
"step": 48
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.909090909090908,
|
| 252 |
+
"eval_loss": 0.6169032454490662,
|
| 253 |
+
"eval_runtime": 25.0151,
|
| 254 |
+
"eval_samples_per_second": 0.959,
|
| 255 |
+
"eval_steps_per_second": 0.959,
|
| 256 |
+
"step": 49
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 9.090909090909092,
|
| 260 |
+
"grad_norm": 0.44522011280059814,
|
| 261 |
+
"learning_rate": 9.698463103929542e-05,
|
| 262 |
+
"loss": 0.3687,
|
| 263 |
+
"step": 50
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.454545454545455,
|
| 267 |
+
"grad_norm": 0.40375861525535583,
|
| 268 |
+
"learning_rate": 9.648882429441257e-05,
|
| 269 |
+
"loss": 0.2941,
|
| 270 |
+
"step": 52
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.818181818181818,
|
| 274 |
+
"grad_norm": 0.480668842792511,
|
| 275 |
+
"learning_rate": 9.595676696276172e-05,
|
| 276 |
+
"loss": 0.2923,
|
| 277 |
+
"step": 54
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 10.0,
|
| 281 |
+
"eval_loss": 0.6638128161430359,
|
| 282 |
+
"eval_runtime": 25.0073,
|
| 283 |
+
"eval_samples_per_second": 0.96,
|
| 284 |
+
"eval_steps_per_second": 0.96,
|
| 285 |
+
"step": 55
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 10.181818181818182,
|
| 289 |
+
"grad_norm": 0.44117558002471924,
|
| 290 |
+
"learning_rate": 9.538887392664544e-05,
|
| 291 |
+
"loss": 0.246,
|
| 292 |
+
"step": 56
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.545454545454545,
|
| 296 |
+
"grad_norm": 0.492409884929657,
|
| 297 |
+
"learning_rate": 9.478558801197065e-05,
|
| 298 |
+
"loss": 0.2068,
|
| 299 |
+
"step": 58
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.909090909090908,
|
| 303 |
+
"grad_norm": 0.6178811192512512,
|
| 304 |
+
"learning_rate": 9.414737964294636e-05,
|
| 305 |
+
"loss": 0.2063,
|
| 306 |
+
"step": 60
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 10.909090909090908,
|
| 310 |
+
"eval_loss": 0.7678647041320801,
|
| 311 |
+
"eval_runtime": 25.0103,
|
| 312 |
+
"eval_samples_per_second": 0.96,
|
| 313 |
+
"eval_steps_per_second": 0.96,
|
| 314 |
+
"step": 60
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 11.272727272727273,
|
| 318 |
+
"grad_norm": 0.5061853528022766,
|
| 319 |
+
"learning_rate": 9.347474647526095e-05,
|
| 320 |
+
"loss": 0.1539,
|
| 321 |
+
"step": 62
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 11.636363636363637,
|
| 325 |
+
"grad_norm": 0.44892534613609314,
|
| 326 |
+
"learning_rate": 9.276821300802534e-05,
|
| 327 |
+
"loss": 0.1254,
|
| 328 |
+
"step": 64
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 12.0,
|
| 332 |
+
"grad_norm": 0.643571674823761,
|
| 333 |
+
"learning_rate": 9.202833017478422e-05,
|
| 334 |
+
"loss": 0.1206,
|
| 335 |
+
"step": 66
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 12.0,
|
| 339 |
+
"eval_loss": 0.858501136302948,
|
| 340 |
+
"eval_runtime": 25.0147,
|
| 341 |
+
"eval_samples_per_second": 0.959,
|
| 342 |
+
"eval_steps_per_second": 0.959,
|
| 343 |
+
"step": 66
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 12.363636363636363,
|
| 347 |
+
"grad_norm": 0.45422667264938354,
|
| 348 |
+
"learning_rate": 9.125567491391476e-05,
|
| 349 |
+
"loss": 0.072,
|
| 350 |
+
"step": 68
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.727272727272727,
|
| 354 |
+
"grad_norm": 0.5882347226142883,
|
| 355 |
+
"learning_rate": 9.045084971874738e-05,
|
| 356 |
+
"loss": 0.0643,
|
| 357 |
+
"step": 70
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 12.909090909090908,
|
| 361 |
+
"eval_loss": 1.0240541696548462,
|
| 362 |
+
"eval_runtime": 25.0111,
|
| 363 |
+
"eval_samples_per_second": 0.96,
|
| 364 |
+
"eval_steps_per_second": 0.96,
|
| 365 |
+
"step": 71
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 13.090909090909092,
|
| 369 |
+
"grad_norm": 0.4740930199623108,
|
| 370 |
+
"learning_rate": 8.961448216775954e-05,
|
| 371 |
+
"loss": 0.0645,
|
| 372 |
+
"step": 72
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.454545454545455,
|
| 376 |
+
"grad_norm": 0.4891839325428009,
|
| 377 |
+
"learning_rate": 8.874722443520899e-05,
|
| 378 |
+
"loss": 0.0418,
|
| 379 |
+
"step": 74
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 13.818181818181818,
|
| 383 |
+
"grad_norm": 0.6079157590866089,
|
| 384 |
+
"learning_rate": 8.784975278258783e-05,
|
| 385 |
+
"loss": 0.0373,
|
| 386 |
+
"step": 76
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 14.0,
|
| 390 |
+
"eval_loss": 1.0912126302719116,
|
| 391 |
+
"eval_runtime": 25.0178,
|
| 392 |
+
"eval_samples_per_second": 0.959,
|
| 393 |
+
"eval_steps_per_second": 0.959,
|
| 394 |
+
"step": 77
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 14.0,
|
| 398 |
+
"step": 77,
|
| 399 |
+
"total_flos": 1.219390286677934e+17,
|
| 400 |
+
"train_loss": 0.4532604143023491,
|
| 401 |
+
"train_runtime": 4348.0981,
|
| 402 |
+
"train_samples_per_second": 1.012,
|
| 403 |
+
"train_steps_per_second": 0.057
|
| 404 |
+
}
|
| 405 |
+
],
|
| 406 |
+
"logging_steps": 2,
|
| 407 |
+
"max_steps": 250,
|
| 408 |
+
"num_input_tokens_seen": 0,
|
| 409 |
+
"num_train_epochs": 50,
|
| 410 |
+
"save_steps": 25,
|
| 411 |
+
"stateful_callbacks": {
|
| 412 |
+
"EarlyStoppingCallback": {
|
| 413 |
+
"args": {
|
| 414 |
+
"early_stopping_patience": 7,
|
| 415 |
+
"early_stopping_threshold": 0.0
|
| 416 |
+
},
|
| 417 |
+
"attributes": {
|
| 418 |
+
"early_stopping_patience_counter": 0
|
| 419 |
+
}
|
| 420 |
+
},
|
| 421 |
+
"TrainerControl": {
|
| 422 |
+
"args": {
|
| 423 |
+
"should_epoch_stop": false,
|
| 424 |
+
"should_evaluate": false,
|
| 425 |
+
"should_log": false,
|
| 426 |
+
"should_save": true,
|
| 427 |
+
"should_training_stop": true
|
| 428 |
+
},
|
| 429 |
+
"attributes": {}
|
| 430 |
+
}
|
| 431 |
+
},
|
| 432 |
+
"total_flos": 1.219390286677934e+17,
|
| 433 |
+
"train_batch_size": 1,
|
| 434 |
+
"trial_name": null,
|
| 435 |
+
"trial_params": null
|
| 436 |
+
}
|