Model save
Browse files- README.md +68 -0
- all_results.json +9 -0
- generation_config.json +7 -0
- train_results.json +9 -0
- trainer_state.json +603 -0
README.md
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: gemma
|
| 4 |
+
base_model: google/gemma-7b
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: gemma7b-gpt4o_1k_summarize-fft
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# gemma7b-gpt4o_1k_summarize-fft
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 6.4970
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0003
|
| 43 |
+
- train_batch_size: 2
|
| 44 |
+
- eval_batch_size: 2
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 8
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 32
|
| 50 |
+
- total_eval_batch_size: 16
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 54 |
+
- num_epochs: 1
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 1.9687 | 1.0 | 392 | 6.4970 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- Transformers 4.45.1
|
| 66 |
+
- Pytorch 2.4.1+cu121
|
| 67 |
+
- Datasets 3.0.1
|
| 68 |
+
- Tokenizers 0.20.0
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 13478110691328.0,
|
| 4 |
+
"train_loss": 7.205524507833987,
|
| 5 |
+
"train_runtime": 9216.1272,
|
| 6 |
+
"train_samples": 115376,
|
| 7 |
+
"train_samples_per_second": 1.36,
|
| 8 |
+
"train_steps_per_second": 0.043
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.45.1"
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 13478110691328.0,
|
| 4 |
+
"train_loss": 7.205524507833987,
|
| 5 |
+
"train_runtime": 9216.1272,
|
| 6 |
+
"train_samples": 115376,
|
| 7 |
+
"train_samples_per_second": 1.36,
|
| 8 |
+
"train_steps_per_second": 0.043
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,603 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 392,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.002551020408163265,
|
| 13 |
+
"grad_norm": 14000.175878727785,
|
| 14 |
+
"learning_rate": 7.499999999999999e-06,
|
| 15 |
+
"loss": 48.0695,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.012755102040816327,
|
| 20 |
+
"grad_norm": 3007.3104608903254,
|
| 21 |
+
"learning_rate": 3.75e-05,
|
| 22 |
+
"loss": 32.7037,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.025510204081632654,
|
| 27 |
+
"grad_norm": 4753.9239460531635,
|
| 28 |
+
"learning_rate": 7.5e-05,
|
| 29 |
+
"loss": 27.6128,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.03826530612244898,
|
| 34 |
+
"grad_norm": 7672.954437161885,
|
| 35 |
+
"learning_rate": 0.0001125,
|
| 36 |
+
"loss": 118.424,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05102040816326531,
|
| 41 |
+
"grad_norm": 2658.2281406370403,
|
| 42 |
+
"learning_rate": 0.00015,
|
| 43 |
+
"loss": 36.7995,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06377551020408163,
|
| 48 |
+
"grad_norm": 301.82927722978826,
|
| 49 |
+
"learning_rate": 0.00018749999999999998,
|
| 50 |
+
"loss": 25.8987,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07653061224489796,
|
| 55 |
+
"grad_norm": 261.2743806851815,
|
| 56 |
+
"learning_rate": 0.000225,
|
| 57 |
+
"loss": 19.7068,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08928571428571429,
|
| 62 |
+
"grad_norm": 170.80865284544745,
|
| 63 |
+
"learning_rate": 0.0002625,
|
| 64 |
+
"loss": 16.2639,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.10204081632653061,
|
| 69 |
+
"grad_norm": 371.9839372489695,
|
| 70 |
+
"learning_rate": 0.0003,
|
| 71 |
+
"loss": 14.8666,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.11479591836734694,
|
| 76 |
+
"grad_norm": 55.46043538057413,
|
| 77 |
+
"learning_rate": 0.00029985067118498503,
|
| 78 |
+
"loss": 13.883,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.12755102040816327,
|
| 83 |
+
"grad_norm": 61.65936269712346,
|
| 84 |
+
"learning_rate": 0.00029940298206120687,
|
| 85 |
+
"loss": 11.3142,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.14030612244897958,
|
| 90 |
+
"grad_norm": 49.78316202658068,
|
| 91 |
+
"learning_rate": 0.0002986578240004834,
|
| 92 |
+
"loss": 10.2446,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.15306122448979592,
|
| 97 |
+
"grad_norm": 95.22883433955214,
|
| 98 |
+
"learning_rate": 0.0002976166806504174,
|
| 99 |
+
"loss": 9.2125,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.16581632653061223,
|
| 104 |
+
"grad_norm": 87.6593544514716,
|
| 105 |
+
"learning_rate": 0.00029628162498037856,
|
| 106 |
+
"loss": 8.0024,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.17857142857142858,
|
| 111 |
+
"grad_norm": 29.457366532379673,
|
| 112 |
+
"learning_rate": 0.000294655315154116,
|
| 113 |
+
"loss": 7.0794,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.1913265306122449,
|
| 118 |
+
"grad_norm": 23.54866864598235,
|
| 119 |
+
"learning_rate": 0.0002927409892372191,
|
| 120 |
+
"loss": 6.428,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.20408163265306123,
|
| 125 |
+
"grad_norm": 14.487277559435018,
|
| 126 |
+
"learning_rate": 0.00029054245874996426,
|
| 127 |
+
"loss": 6.0644,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.21683673469387754,
|
| 132 |
+
"grad_norm": 11.612016698553127,
|
| 133 |
+
"learning_rate": 0.00028806410107838377,
|
| 134 |
+
"loss": 5.9737,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.22959183673469388,
|
| 139 |
+
"grad_norm": 22.24847439069348,
|
| 140 |
+
"learning_rate": 0.00028531085075866815,
|
| 141 |
+
"loss": 5.6293,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.2423469387755102,
|
| 146 |
+
"grad_norm": 25.32273350126865,
|
| 147 |
+
"learning_rate": 0.0002822881896522532,
|
| 148 |
+
"loss": 5.4368,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.25510204081632654,
|
| 153 |
+
"grad_norm": 29.73658242449881,
|
| 154 |
+
"learning_rate": 0.000279002136031155,
|
| 155 |
+
"loss": 5.2007,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.26785714285714285,
|
| 160 |
+
"grad_norm": 17.009329865460675,
|
| 161 |
+
"learning_rate": 0.0002754592325952845,
|
| 162 |
+
"loss": 4.9784,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.28061224489795916,
|
| 167 |
+
"grad_norm": 17.65449666930728,
|
| 168 |
+
"learning_rate": 0.00027166653344559883,
|
| 169 |
+
"loss": 5.1761,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.29336734693877553,
|
| 174 |
+
"grad_norm": 25.60749450078011,
|
| 175 |
+
"learning_rate": 0.0002676315900390278,
|
| 176 |
+
"loss": 4.981,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.30612244897959184,
|
| 181 |
+
"grad_norm": 19.37666875758449,
|
| 182 |
+
"learning_rate": 0.00026336243615313873,
|
| 183 |
+
"loss": 4.883,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.31887755102040816,
|
| 188 |
+
"grad_norm": 25.223445103235044,
|
| 189 |
+
"learning_rate": 0.00025886757189047695,
|
| 190 |
+
"loss": 4.6906,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.33163265306122447,
|
| 195 |
+
"grad_norm": 24.827221258799884,
|
| 196 |
+
"learning_rate": 0.0002541559467544291,
|
| 197 |
+
"loss": 4.5962,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.34438775510204084,
|
| 202 |
+
"grad_norm": 19.172749908950284,
|
| 203 |
+
"learning_rate": 0.00024923694183030657,
|
| 204 |
+
"loss": 4.4971,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.35714285714285715,
|
| 209 |
+
"grad_norm": 17.803290735515485,
|
| 210 |
+
"learning_rate": 0.0002441203511071278,
|
| 211 |
+
"loss": 4.3923,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.36989795918367346,
|
| 216 |
+
"grad_norm": 15.152588588737128,
|
| 217 |
+
"learning_rate": 0.00023881636197728743,
|
| 218 |
+
"loss": 4.2507,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3826530612244898,
|
| 223 |
+
"grad_norm": 28.266308285286772,
|
| 224 |
+
"learning_rate": 0.0002333355349529403,
|
| 225 |
+
"loss": 4.1071,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.39540816326530615,
|
| 230 |
+
"grad_norm": 13.058293075351123,
|
| 231 |
+
"learning_rate": 0.00022768878263948445,
|
| 232 |
+
"loss": 4.0515,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.40816326530612246,
|
| 237 |
+
"grad_norm": 14.87484760188348,
|
| 238 |
+
"learning_rate": 0.00022188734800800852,
|
| 239 |
+
"loss": 3.9059,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.42091836734693877,
|
| 244 |
+
"grad_norm": 12.57655967452742,
|
| 245 |
+
"learning_rate": 0.00021594278200996457,
|
| 246 |
+
"loss": 3.876,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.4336734693877551,
|
| 251 |
+
"grad_norm": 18.644576504823885,
|
| 252 |
+
"learning_rate": 0.00020986692057863607,
|
| 253 |
+
"loss": 3.703,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.44642857142857145,
|
| 258 |
+
"grad_norm": 6.3705323701789265,
|
| 259 |
+
"learning_rate": 0.000203671861063193,
|
| 260 |
+
"loss": 3.6606,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.45918367346938777,
|
| 265 |
+
"grad_norm": 9.945740907460738,
|
| 266 |
+
"learning_rate": 0.00019736993814225374,
|
| 267 |
+
"loss": 3.6781,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4719387755102041,
|
| 272 |
+
"grad_norm": 5.28597104720817,
|
| 273 |
+
"learning_rate": 0.00019097369926491297,
|
| 274 |
+
"loss": 3.5963,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.4846938775510204,
|
| 279 |
+
"grad_norm": 5.570474507029548,
|
| 280 |
+
"learning_rate": 0.00018449587966813208,
|
| 281 |
+
"loss": 3.5206,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.49744897959183676,
|
| 286 |
+
"grad_norm": 5.769299754282368,
|
| 287 |
+
"learning_rate": 0.00017794937702023467,
|
| 288 |
+
"loss": 3.3751,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5102040816326531,
|
| 293 |
+
"grad_norm": 7.514106159557929,
|
| 294 |
+
"learning_rate": 0.00017134722574099276,
|
| 295 |
+
"loss": 3.3315,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5229591836734694,
|
| 300 |
+
"grad_norm": 5.467040990097222,
|
| 301 |
+
"learning_rate": 0.0001647025710494341,
|
| 302 |
+
"loss": 3.2297,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5357142857142857,
|
| 307 |
+
"grad_norm": 7.752282985206345,
|
| 308 |
+
"learning_rate": 0.00015802864279104223,
|
| 309 |
+
"loss": 3.1714,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.548469387755102,
|
| 314 |
+
"grad_norm": 4.902568385413757,
|
| 315 |
+
"learning_rate": 0.0001513387290964616,
|
| 316 |
+
"loss": 3.076,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5612244897959183,
|
| 321 |
+
"grad_norm": 8.809433652257214,
|
| 322 |
+
"learning_rate": 0.00014464614992415294,
|
| 323 |
+
"loss": 3.0378,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5739795918367347,
|
| 328 |
+
"grad_norm": 5.62075198192736,
|
| 329 |
+
"learning_rate": 0.00013796423053967887,
|
| 330 |
+
"loss": 2.946,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5867346938775511,
|
| 335 |
+
"grad_norm": 7.091999165572766,
|
| 336 |
+
"learning_rate": 0.0001313062749844221,
|
| 337 |
+
"loss": 2.9117,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.5994897959183674,
|
| 342 |
+
"grad_norm": 9.664621289964327,
|
| 343 |
+
"learning_rate": 0.00012468553958656264,
|
| 344 |
+
"loss": 2.7779,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6122448979591837,
|
| 349 |
+
"grad_norm": 8.079291607422778,
|
| 350 |
+
"learning_rate": 0.00011811520656705348,
|
| 351 |
+
"loss": 2.7627,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.625,
|
| 356 |
+
"grad_norm": 9.132106023308898,
|
| 357 |
+
"learning_rate": 0.00011160835779314889,
|
| 358 |
+
"loss": 2.7464,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6377551020408163,
|
| 363 |
+
"grad_norm": 12.688497775305251,
|
| 364 |
+
"learning_rate": 0.00010517794873174064,
|
| 365 |
+
"loss": 2.6784,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6505102040816326,
|
| 370 |
+
"grad_norm": 7.508821446542865,
|
| 371 |
+
"learning_rate": 9.883678265436473e-05,
|
| 372 |
+
"loss": 2.6542,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6632653061224489,
|
| 377 |
+
"grad_norm": 7.288150995036509,
|
| 378 |
+
"learning_rate": 9.259748514523653e-05,
|
| 379 |
+
"loss": 2.5675,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6760204081632653,
|
| 384 |
+
"grad_norm": 2.598306498532322,
|
| 385 |
+
"learning_rate": 8.647247896307018e-05,
|
| 386 |
+
"loss": 2.5179,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.6887755102040817,
|
| 391 |
+
"grad_norm": 2.9340097554582947,
|
| 392 |
+
"learning_rate": 8.047395930673417e-05,
|
| 393 |
+
"loss": 2.4673,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.701530612244898,
|
| 398 |
+
"grad_norm": 3.520741961544552,
|
| 399 |
+
"learning_rate": 7.46138695339903e-05,
|
| 400 |
+
"loss": 2.5285,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7142857142857143,
|
| 405 |
+
"grad_norm": 4.635037745202294,
|
| 406 |
+
"learning_rate": 6.890387738166041e-05,
|
| 407 |
+
"loss": 2.3872,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7270408163265306,
|
| 412 |
+
"grad_norm": 3.987422126482862,
|
| 413 |
+
"learning_rate": 6.335535173456914e-05,
|
| 414 |
+
"loss": 2.3409,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7397959183673469,
|
| 419 |
+
"grad_norm": 3.0549621746230122,
|
| 420 |
+
"learning_rate": 5.79793399895161e-05,
|
| 421 |
+
"loss": 2.3254,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7525510204081632,
|
| 426 |
+
"grad_norm": 2.527266957929868,
|
| 427 |
+
"learning_rate": 5.278654605934644e-05,
|
| 428 |
+
"loss": 2.3429,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7653061224489796,
|
| 433 |
+
"grad_norm": 3.9761126497464083,
|
| 434 |
+
"learning_rate": 4.778730906091632e-05,
|
| 435 |
+
"loss": 2.2348,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.7780612244897959,
|
| 440 |
+
"grad_norm": 2.498797778043332,
|
| 441 |
+
"learning_rate": 4.2991582729385174e-05,
|
| 442 |
+
"loss": 2.2421,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.7908163265306123,
|
| 447 |
+
"grad_norm": 2.5462712929101894,
|
| 448 |
+
"learning_rate": 3.840891559982256e-05,
|
| 449 |
+
"loss": 2.2751,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8035714285714286,
|
| 454 |
+
"grad_norm": 2.307072702085812,
|
| 455 |
+
"learning_rate": 3.404843199558945e-05,
|
| 456 |
+
"loss": 2.1867,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8163265306122449,
|
| 461 |
+
"grad_norm": 5.263562323908358,
|
| 462 |
+
"learning_rate": 2.9918813861345952e-05,
|
| 463 |
+
"loss": 2.181,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8290816326530612,
|
| 468 |
+
"grad_norm": 2.0011040214883575,
|
| 469 |
+
"learning_rate": 2.6028283476858038e-05,
|
| 470 |
+
"loss": 2.1732,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8418367346938775,
|
| 475 |
+
"grad_norm": 2.2471501177416116,
|
| 476 |
+
"learning_rate": 2.238458708602039e-05,
|
| 477 |
+
"loss": 2.1108,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8545918367346939,
|
| 482 |
+
"grad_norm": 1.66846134386125,
|
| 483 |
+
"learning_rate": 1.8994979473690537e-05,
|
| 484 |
+
"loss": 2.1007,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.8673469387755102,
|
| 489 |
+
"grad_norm": 2.0450866177774976,
|
| 490 |
+
"learning_rate": 1.5866209521043304e-05,
|
| 491 |
+
"loss": 2.0334,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.8801020408163265,
|
| 496 |
+
"grad_norm": 2.1608778408402807,
|
| 497 |
+
"learning_rate": 1.3004506768205226e-05,
|
| 498 |
+
"loss": 2.0735,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.8928571428571429,
|
| 503 |
+
"grad_norm": 2.1504307790627912,
|
| 504 |
+
"learning_rate": 1.0415569010922963e-05,
|
| 505 |
+
"loss": 2.0473,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9056122448979592,
|
| 510 |
+
"grad_norm": 1.6880900496760922,
|
| 511 |
+
"learning_rate": 8.104550955962469e-06,
|
| 512 |
+
"loss": 2.029,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9183673469387755,
|
| 517 |
+
"grad_norm": 1.7900392749998477,
|
| 518 |
+
"learning_rate": 6.076053957825411e-06,
|
| 519 |
+
"loss": 2.0533,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9311224489795918,
|
| 524 |
+
"grad_norm": 1.5178075882516209,
|
| 525 |
+
"learning_rate": 4.334116857218317e-06,
|
| 526 |
+
"loss": 1.9693,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.9438775510204082,
|
| 531 |
+
"grad_norm": 1.6874487405997367,
|
| 532 |
+
"learning_rate": 2.882207939515435e-06,
|
| 533 |
+
"loss": 1.9791,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9566326530612245,
|
| 538 |
+
"grad_norm": 1.4244094850155935,
|
| 539 |
+
"learning_rate": 1.7232180292259369e-06,
|
| 540 |
+
"loss": 1.9566,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.9693877551020408,
|
| 545 |
+
"grad_norm": 1.302432101303747,
|
| 546 |
+
"learning_rate": 8.594547342153979e-07,
|
| 547 |
+
"loss": 1.9451,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.9821428571428571,
|
| 552 |
+
"grad_norm": 1.2049628347309629,
|
| 553 |
+
"learning_rate": 2.926378511411198e-07,
|
| 554 |
+
"loss": 1.9952,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.9948979591836735,
|
| 559 |
+
"grad_norm": 1.2981702353996074,
|
| 560 |
+
"learning_rate": 2.3895941249507665e-08,
|
| 561 |
+
"loss": 1.9687,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.0,
|
| 566 |
+
"eval_loss": 6.496990203857422,
|
| 567 |
+
"eval_runtime": 2.5179,
|
| 568 |
+
"eval_samples_per_second": 3.972,
|
| 569 |
+
"eval_steps_per_second": 0.397,
|
| 570 |
+
"step": 392
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.0,
|
| 574 |
+
"step": 392,
|
| 575 |
+
"total_flos": 13478110691328.0,
|
| 576 |
+
"train_loss": 7.205524507833987,
|
| 577 |
+
"train_runtime": 9216.1272,
|
| 578 |
+
"train_samples_per_second": 1.36,
|
| 579 |
+
"train_steps_per_second": 0.043
|
| 580 |
+
}
|
| 581 |
+
],
|
| 582 |
+
"logging_steps": 5,
|
| 583 |
+
"max_steps": 392,
|
| 584 |
+
"num_input_tokens_seen": 0,
|
| 585 |
+
"num_train_epochs": 1,
|
| 586 |
+
"save_steps": 100,
|
| 587 |
+
"stateful_callbacks": {
|
| 588 |
+
"TrainerControl": {
|
| 589 |
+
"args": {
|
| 590 |
+
"should_epoch_stop": false,
|
| 591 |
+
"should_evaluate": false,
|
| 592 |
+
"should_log": false,
|
| 593 |
+
"should_save": true,
|
| 594 |
+
"should_training_stop": true
|
| 595 |
+
},
|
| 596 |
+
"attributes": {}
|
| 597 |
+
}
|
| 598 |
+
},
|
| 599 |
+
"total_flos": 13478110691328.0,
|
| 600 |
+
"train_batch_size": 2,
|
| 601 |
+
"trial_name": null,
|
| 602 |
+
"trial_params": null
|
| 603 |
+
}
|