Commit
•
22da792
1
Parent(s):
79bfe94
Model save
Browse files- README.md +77 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +547 -0
README.md
ADDED
|
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.3
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: apache-2.0
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Mistral-7B_task-1_180-samples_config-2_full
|
| 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 |
+
# Mistral-7B_task-1_180-samples_config-2_full
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.4892
|
| 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 |
+
| 1.9695 | 0.9412 | 8 | 1.8462 |
|
| 57 |
+
| 1.4658 | 2.0 | 17 | 1.3229 |
|
| 58 |
+
| 1.0873 | 2.9412 | 25 | 1.0281 |
|
| 59 |
+
| 0.8963 | 4.0 | 34 | 0.9578 |
|
| 60 |
+
| 0.8173 | 4.9412 | 42 | 0.9242 |
|
| 61 |
+
| 0.7207 | 6.0 | 51 | 0.9159 |
|
| 62 |
+
| 0.5468 | 6.9412 | 59 | 0.9483 |
|
| 63 |
+
| 0.4071 | 8.0 | 68 | 1.0524 |
|
| 64 |
+
| 0.2885 | 8.9412 | 76 | 1.2125 |
|
| 65 |
+
| 0.2064 | 10.0 | 85 | 1.2909 |
|
| 66 |
+
| 0.1585 | 10.9412 | 93 | 1.3461 |
|
| 67 |
+
| 0.125 | 12.0 | 102 | 1.3770 |
|
| 68 |
+
| 0.1121 | 12.9412 | 110 | 1.4892 |
|
| 69 |
+
|
| 70 |
+
|
| 71 |
+
### Framework versions
|
| 72 |
+
|
| 73 |
+
- PEFT 0.12.0
|
| 74 |
+
- Transformers 4.44.0
|
| 75 |
+
- Pytorch 2.1.2+cu121
|
| 76 |
+
- Datasets 2.20.0
|
| 77 |
+
- 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 sha256:9042a6e4b3609b8865a7e8e715bf42d8576042632887cfd356f088ad6ec701f7
|
| 3 |
size 167832240
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.941176470588236,
|
| 3 |
+
"total_flos": 2.772769687207936e+16,
|
| 4 |
+
"train_loss": 0.7063802681524646,
|
| 5 |
+
"train_runtime": 2094.2057,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 3.247,
|
| 8 |
+
"train_steps_per_second": 0.191
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.941176470588236,
|
| 3 |
+
"total_flos": 2.772769687207936e+16,
|
| 4 |
+
"train_loss": 0.7063802681524646,
|
| 5 |
+
"train_runtime": 2094.2057,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 3.247,
|
| 8 |
+
"train_steps_per_second": 0.191
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,547 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9159461259841919,
|
| 3 |
+
"best_model_checkpoint": "data/Mistral-7B_task-1_180-samples_config-2_full/checkpoint-51",
|
| 4 |
+
"epoch": 12.941176470588236,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 110,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11764705882352941,
|
| 13 |
+
"grad_norm": 2.5466060638427734,
|
| 14 |
+
"learning_rate": 2.5e-06,
|
| 15 |
+
"loss": 2.075,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.23529411764705882,
|
| 20 |
+
"grad_norm": 2.510213851928711,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 2.0283,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.47058823529411764,
|
| 27 |
+
"grad_norm": 2.568328380584717,
|
| 28 |
+
"learning_rate": 1e-05,
|
| 29 |
+
"loss": 2.0754,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.7058823529411765,
|
| 34 |
+
"grad_norm": 2.496351957321167,
|
| 35 |
+
"learning_rate": 1.5e-05,
|
| 36 |
+
"loss": 2.0418,
|
| 37 |
+
"step": 6
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.9411764705882353,
|
| 41 |
+
"grad_norm": 2.0648036003112793,
|
| 42 |
+
"learning_rate": 2e-05,
|
| 43 |
+
"loss": 1.9695,
|
| 44 |
+
"step": 8
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.9411764705882353,
|
| 48 |
+
"eval_loss": 1.8462496995925903,
|
| 49 |
+
"eval_runtime": 14.5686,
|
| 50 |
+
"eval_samples_per_second": 2.471,
|
| 51 |
+
"eval_steps_per_second": 2.471,
|
| 52 |
+
"step": 8
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.1764705882352942,
|
| 56 |
+
"grad_norm": 1.5974884033203125,
|
| 57 |
+
"learning_rate": 2.5e-05,
|
| 58 |
+
"loss": 1.8101,
|
| 59 |
+
"step": 10
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.4117647058823528,
|
| 63 |
+
"grad_norm": 1.2964701652526855,
|
| 64 |
+
"learning_rate": 3e-05,
|
| 65 |
+
"loss": 1.7043,
|
| 66 |
+
"step": 12
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.6470588235294117,
|
| 70 |
+
"grad_norm": 1.2164214849472046,
|
| 71 |
+
"learning_rate": 3.5e-05,
|
| 72 |
+
"loss": 1.6112,
|
| 73 |
+
"step": 14
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.8823529411764706,
|
| 77 |
+
"grad_norm": 1.3542946577072144,
|
| 78 |
+
"learning_rate": 4e-05,
|
| 79 |
+
"loss": 1.4658,
|
| 80 |
+
"step": 16
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.0,
|
| 84 |
+
"eval_loss": 1.3228856325149536,
|
| 85 |
+
"eval_runtime": 14.5474,
|
| 86 |
+
"eval_samples_per_second": 2.475,
|
| 87 |
+
"eval_steps_per_second": 2.475,
|
| 88 |
+
"step": 17
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.1176470588235294,
|
| 92 |
+
"grad_norm": 1.3138030767440796,
|
| 93 |
+
"learning_rate": 4.5e-05,
|
| 94 |
+
"loss": 1.3523,
|
| 95 |
+
"step": 18
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.3529411764705883,
|
| 99 |
+
"grad_norm": 1.0838539600372314,
|
| 100 |
+
"learning_rate": 5e-05,
|
| 101 |
+
"loss": 1.2067,
|
| 102 |
+
"step": 20
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 2.588235294117647,
|
| 106 |
+
"grad_norm": 1.0649068355560303,
|
| 107 |
+
"learning_rate": 5.500000000000001e-05,
|
| 108 |
+
"loss": 1.1172,
|
| 109 |
+
"step": 22
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 2.8235294117647056,
|
| 113 |
+
"grad_norm": 0.7390829920768738,
|
| 114 |
+
"learning_rate": 6e-05,
|
| 115 |
+
"loss": 1.0873,
|
| 116 |
+
"step": 24
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.9411764705882355,
|
| 120 |
+
"eval_loss": 1.0281312465667725,
|
| 121 |
+
"eval_runtime": 14.5775,
|
| 122 |
+
"eval_samples_per_second": 2.47,
|
| 123 |
+
"eval_steps_per_second": 2.47,
|
| 124 |
+
"step": 25
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 3.0588235294117645,
|
| 128 |
+
"grad_norm": 0.6466070413589478,
|
| 129 |
+
"learning_rate": 6.500000000000001e-05,
|
| 130 |
+
"loss": 0.9773,
|
| 131 |
+
"step": 26
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.2941176470588234,
|
| 135 |
+
"grad_norm": 0.6372568607330322,
|
| 136 |
+
"learning_rate": 7e-05,
|
| 137 |
+
"loss": 1.0028,
|
| 138 |
+
"step": 28
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 3.5294117647058822,
|
| 142 |
+
"grad_norm": 0.6881040334701538,
|
| 143 |
+
"learning_rate": 7.500000000000001e-05,
|
| 144 |
+
"loss": 0.9714,
|
| 145 |
+
"step": 30
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 3.764705882352941,
|
| 149 |
+
"grad_norm": 0.5336817502975464,
|
| 150 |
+
"learning_rate": 8e-05,
|
| 151 |
+
"loss": 0.9352,
|
| 152 |
+
"step": 32
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.0,
|
| 156 |
+
"grad_norm": 0.5645790100097656,
|
| 157 |
+
"learning_rate": 8.5e-05,
|
| 158 |
+
"loss": 0.8963,
|
| 159 |
+
"step": 34
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 4.0,
|
| 163 |
+
"eval_loss": 0.9577707648277283,
|
| 164 |
+
"eval_runtime": 14.5662,
|
| 165 |
+
"eval_samples_per_second": 2.471,
|
| 166 |
+
"eval_steps_per_second": 2.471,
|
| 167 |
+
"step": 34
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 4.235294117647059,
|
| 171 |
+
"grad_norm": 0.5516751408576965,
|
| 172 |
+
"learning_rate": 9e-05,
|
| 173 |
+
"loss": 0.8726,
|
| 174 |
+
"step": 36
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 4.470588235294118,
|
| 178 |
+
"grad_norm": 0.6818626523017883,
|
| 179 |
+
"learning_rate": 9.5e-05,
|
| 180 |
+
"loss": 0.8671,
|
| 181 |
+
"step": 38
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 4.705882352941177,
|
| 185 |
+
"grad_norm": 0.6951989531517029,
|
| 186 |
+
"learning_rate": 0.0001,
|
| 187 |
+
"loss": 0.8326,
|
| 188 |
+
"step": 40
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.9411764705882355,
|
| 192 |
+
"grad_norm": 0.6834418773651123,
|
| 193 |
+
"learning_rate": 9.999238475781957e-05,
|
| 194 |
+
"loss": 0.8173,
|
| 195 |
+
"step": 42
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 4.9411764705882355,
|
| 199 |
+
"eval_loss": 0.9242374897003174,
|
| 200 |
+
"eval_runtime": 14.5691,
|
| 201 |
+
"eval_samples_per_second": 2.471,
|
| 202 |
+
"eval_steps_per_second": 2.471,
|
| 203 |
+
"step": 42
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.176470588235294,
|
| 207 |
+
"grad_norm": 0.6817136406898499,
|
| 208 |
+
"learning_rate": 9.99695413509548e-05,
|
| 209 |
+
"loss": 0.7695,
|
| 210 |
+
"step": 44
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.411764705882353,
|
| 214 |
+
"grad_norm": 0.6675918698310852,
|
| 215 |
+
"learning_rate": 9.99314767377287e-05,
|
| 216 |
+
"loss": 0.7399,
|
| 217 |
+
"step": 46
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.647058823529412,
|
| 221 |
+
"grad_norm": 0.8630544543266296,
|
| 222 |
+
"learning_rate": 9.987820251299122e-05,
|
| 223 |
+
"loss": 0.6773,
|
| 224 |
+
"step": 48
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 5.882352941176471,
|
| 228 |
+
"grad_norm": 0.8795169591903687,
|
| 229 |
+
"learning_rate": 9.980973490458728e-05,
|
| 230 |
+
"loss": 0.7207,
|
| 231 |
+
"step": 50
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.0,
|
| 235 |
+
"eval_loss": 0.9159461259841919,
|
| 236 |
+
"eval_runtime": 14.5733,
|
| 237 |
+
"eval_samples_per_second": 2.47,
|
| 238 |
+
"eval_steps_per_second": 2.47,
|
| 239 |
+
"step": 51
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 6.117647058823529,
|
| 243 |
+
"grad_norm": 0.7988505363464355,
|
| 244 |
+
"learning_rate": 9.972609476841367e-05,
|
| 245 |
+
"loss": 0.6305,
|
| 246 |
+
"step": 52
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 6.352941176470588,
|
| 250 |
+
"grad_norm": 0.8636096715927124,
|
| 251 |
+
"learning_rate": 9.962730758206611e-05,
|
| 252 |
+
"loss": 0.5937,
|
| 253 |
+
"step": 54
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 6.588235294117647,
|
| 257 |
+
"grad_norm": 1.080081582069397,
|
| 258 |
+
"learning_rate": 9.951340343707852e-05,
|
| 259 |
+
"loss": 0.5753,
|
| 260 |
+
"step": 56
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 6.823529411764706,
|
| 264 |
+
"grad_norm": 1.0529221296310425,
|
| 265 |
+
"learning_rate": 9.938441702975689e-05,
|
| 266 |
+
"loss": 0.5468,
|
| 267 |
+
"step": 58
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.9411764705882355,
|
| 271 |
+
"eval_loss": 0.9483461380004883,
|
| 272 |
+
"eval_runtime": 14.5621,
|
| 273 |
+
"eval_samples_per_second": 2.472,
|
| 274 |
+
"eval_steps_per_second": 2.472,
|
| 275 |
+
"step": 59
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.0588235294117645,
|
| 279 |
+
"grad_norm": 0.9509832262992859,
|
| 280 |
+
"learning_rate": 9.924038765061042e-05,
|
| 281 |
+
"loss": 0.5346,
|
| 282 |
+
"step": 60
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.294117647058823,
|
| 286 |
+
"grad_norm": 0.9746370315551758,
|
| 287 |
+
"learning_rate": 9.908135917238321e-05,
|
| 288 |
+
"loss": 0.4425,
|
| 289 |
+
"step": 62
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 7.529411764705882,
|
| 293 |
+
"grad_norm": 1.384165644645691,
|
| 294 |
+
"learning_rate": 9.890738003669029e-05,
|
| 295 |
+
"loss": 0.4108,
|
| 296 |
+
"step": 64
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.764705882352941,
|
| 300 |
+
"grad_norm": 1.441620945930481,
|
| 301 |
+
"learning_rate": 9.871850323926177e-05,
|
| 302 |
+
"loss": 0.3867,
|
| 303 |
+
"step": 66
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 8.0,
|
| 307 |
+
"grad_norm": 1.274566650390625,
|
| 308 |
+
"learning_rate": 9.851478631379982e-05,
|
| 309 |
+
"loss": 0.4071,
|
| 310 |
+
"step": 68
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 8.0,
|
| 314 |
+
"eval_loss": 1.0524382591247559,
|
| 315 |
+
"eval_runtime": 14.5602,
|
| 316 |
+
"eval_samples_per_second": 2.472,
|
| 317 |
+
"eval_steps_per_second": 2.472,
|
| 318 |
+
"step": 68
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 8.235294117647058,
|
| 322 |
+
"grad_norm": 1.0214779376983643,
|
| 323 |
+
"learning_rate": 9.829629131445342e-05,
|
| 324 |
+
"loss": 0.3106,
|
| 325 |
+
"step": 70
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 8.470588235294118,
|
| 329 |
+
"grad_norm": 1.312712550163269,
|
| 330 |
+
"learning_rate": 9.806308479691595e-05,
|
| 331 |
+
"loss": 0.303,
|
| 332 |
+
"step": 72
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 8.705882352941176,
|
| 336 |
+
"grad_norm": 1.6440112590789795,
|
| 337 |
+
"learning_rate": 9.781523779815179e-05,
|
| 338 |
+
"loss": 0.282,
|
| 339 |
+
"step": 74
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 8.941176470588236,
|
| 343 |
+
"grad_norm": 1.8456093072891235,
|
| 344 |
+
"learning_rate": 9.755282581475769e-05,
|
| 345 |
+
"loss": 0.2885,
|
| 346 |
+
"step": 76
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 8.941176470588236,
|
| 350 |
+
"eval_loss": 1.212545394897461,
|
| 351 |
+
"eval_runtime": 14.5543,
|
| 352 |
+
"eval_samples_per_second": 2.473,
|
| 353 |
+
"eval_steps_per_second": 2.473,
|
| 354 |
+
"step": 76
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 9.176470588235293,
|
| 358 |
+
"grad_norm": 1.2559385299682617,
|
| 359 |
+
"learning_rate": 9.727592877996585e-05,
|
| 360 |
+
"loss": 0.2313,
|
| 361 |
+
"step": 78
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 9.411764705882353,
|
| 365 |
+
"grad_norm": 1.2266454696655273,
|
| 366 |
+
"learning_rate": 9.698463103929542e-05,
|
| 367 |
+
"loss": 0.2054,
|
| 368 |
+
"step": 80
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 9.647058823529411,
|
| 372 |
+
"grad_norm": 1.3700865507125854,
|
| 373 |
+
"learning_rate": 9.667902132486009e-05,
|
| 374 |
+
"loss": 0.2041,
|
| 375 |
+
"step": 82
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 9.882352941176471,
|
| 379 |
+
"grad_norm": 1.4827758073806763,
|
| 380 |
+
"learning_rate": 9.635919272833938e-05,
|
| 381 |
+
"loss": 0.2064,
|
| 382 |
+
"step": 84
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 10.0,
|
| 386 |
+
"eval_loss": 1.290916919708252,
|
| 387 |
+
"eval_runtime": 14.5516,
|
| 388 |
+
"eval_samples_per_second": 2.474,
|
| 389 |
+
"eval_steps_per_second": 2.474,
|
| 390 |
+
"step": 85
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 10.117647058823529,
|
| 394 |
+
"grad_norm": 1.2037745714187622,
|
| 395 |
+
"learning_rate": 9.602524267262203e-05,
|
| 396 |
+
"loss": 0.1821,
|
| 397 |
+
"step": 86
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 10.352941176470589,
|
| 401 |
+
"grad_norm": 1.2108136415481567,
|
| 402 |
+
"learning_rate": 9.567727288213005e-05,
|
| 403 |
+
"loss": 0.1467,
|
| 404 |
+
"step": 88
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 10.588235294117647,
|
| 408 |
+
"grad_norm": 1.2231992483139038,
|
| 409 |
+
"learning_rate": 9.53153893518325e-05,
|
| 410 |
+
"loss": 0.147,
|
| 411 |
+
"step": 90
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 10.823529411764707,
|
| 415 |
+
"grad_norm": 1.1290245056152344,
|
| 416 |
+
"learning_rate": 9.493970231495835e-05,
|
| 417 |
+
"loss": 0.1585,
|
| 418 |
+
"step": 92
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 10.941176470588236,
|
| 422 |
+
"eval_loss": 1.3460688591003418,
|
| 423 |
+
"eval_runtime": 14.5514,
|
| 424 |
+
"eval_samples_per_second": 2.474,
|
| 425 |
+
"eval_steps_per_second": 2.474,
|
| 426 |
+
"step": 93
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 11.058823529411764,
|
| 430 |
+
"grad_norm": 0.832689106464386,
|
| 431 |
+
"learning_rate": 9.45503262094184e-05,
|
| 432 |
+
"loss": 0.1399,
|
| 433 |
+
"step": 94
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 11.294117647058824,
|
| 437 |
+
"grad_norm": 0.9425649642944336,
|
| 438 |
+
"learning_rate": 9.414737964294636e-05,
|
| 439 |
+
"loss": 0.1147,
|
| 440 |
+
"step": 96
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 11.529411764705882,
|
| 444 |
+
"grad_norm": 1.262665033340454,
|
| 445 |
+
"learning_rate": 9.373098535696979e-05,
|
| 446 |
+
"loss": 0.1255,
|
| 447 |
+
"step": 98
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 11.764705882352942,
|
| 451 |
+
"grad_norm": 1.311782717704773,
|
| 452 |
+
"learning_rate": 9.330127018922194e-05,
|
| 453 |
+
"loss": 0.125,
|
| 454 |
+
"step": 100
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 12.0,
|
| 458 |
+
"grad_norm": 0.9994375109672546,
|
| 459 |
+
"learning_rate": 9.285836503510562e-05,
|
| 460 |
+
"loss": 0.125,
|
| 461 |
+
"step": 102
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 12.0,
|
| 465 |
+
"eval_loss": 1.3770337104797363,
|
| 466 |
+
"eval_runtime": 14.5637,
|
| 467 |
+
"eval_samples_per_second": 2.472,
|
| 468 |
+
"eval_steps_per_second": 2.472,
|
| 469 |
+
"step": 102
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 12.235294117647058,
|
| 473 |
+
"grad_norm": 0.99504554271698,
|
| 474 |
+
"learning_rate": 9.24024048078213e-05,
|
| 475 |
+
"loss": 0.1028,
|
| 476 |
+
"step": 104
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 12.470588235294118,
|
| 480 |
+
"grad_norm": 1.4479221105575562,
|
| 481 |
+
"learning_rate": 9.193352839727121e-05,
|
| 482 |
+
"loss": 0.1057,
|
| 483 |
+
"step": 106
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 12.705882352941176,
|
| 487 |
+
"grad_norm": 1.4204106330871582,
|
| 488 |
+
"learning_rate": 9.145187862775209e-05,
|
| 489 |
+
"loss": 0.1063,
|
| 490 |
+
"step": 108
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 12.941176470588236,
|
| 494 |
+
"grad_norm": 1.1615417003631592,
|
| 495 |
+
"learning_rate": 9.09576022144496e-05,
|
| 496 |
+
"loss": 0.1121,
|
| 497 |
+
"step": 110
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 12.941176470588236,
|
| 501 |
+
"eval_loss": 1.4891993999481201,
|
| 502 |
+
"eval_runtime": 14.5634,
|
| 503 |
+
"eval_samples_per_second": 2.472,
|
| 504 |
+
"eval_steps_per_second": 2.472,
|
| 505 |
+
"step": 110
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 12.941176470588236,
|
| 509 |
+
"step": 110,
|
| 510 |
+
"total_flos": 2.772769687207936e+16,
|
| 511 |
+
"train_loss": 0.7063802681524646,
|
| 512 |
+
"train_runtime": 2094.2057,
|
| 513 |
+
"train_samples_per_second": 3.247,
|
| 514 |
+
"train_steps_per_second": 0.191
|
| 515 |
+
}
|
| 516 |
+
],
|
| 517 |
+
"logging_steps": 2,
|
| 518 |
+
"max_steps": 400,
|
| 519 |
+
"num_input_tokens_seen": 0,
|
| 520 |
+
"num_train_epochs": 50,
|
| 521 |
+
"save_steps": 25,
|
| 522 |
+
"stateful_callbacks": {
|
| 523 |
+
"EarlyStoppingCallback": {
|
| 524 |
+
"args": {
|
| 525 |
+
"early_stopping_patience": 7,
|
| 526 |
+
"early_stopping_threshold": 0.0
|
| 527 |
+
},
|
| 528 |
+
"attributes": {
|
| 529 |
+
"early_stopping_patience_counter": 0
|
| 530 |
+
}
|
| 531 |
+
},
|
| 532 |
+
"TrainerControl": {
|
| 533 |
+
"args": {
|
| 534 |
+
"should_epoch_stop": false,
|
| 535 |
+
"should_evaluate": false,
|
| 536 |
+
"should_log": false,
|
| 537 |
+
"should_save": true,
|
| 538 |
+
"should_training_stop": true
|
| 539 |
+
},
|
| 540 |
+
"attributes": {}
|
| 541 |
+
}
|
| 542 |
+
},
|
| 543 |
+
"total_flos": 2.772769687207936e+16,
|
| 544 |
+
"train_batch_size": 1,
|
| 545 |
+
"trial_name": null,
|
| 546 |
+
"trial_params": null
|
| 547 |
+
}
|