Commit
•
7878701
1
Parent(s):
e319c33
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 +1003 -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_180-samples_config-1_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 |
+
# Llama-31-8B_task-2_180-samples_config-1_full
|
| 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: 1.3371
|
| 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: 8
|
| 46 |
+
- total_train_batch_size: 8
|
| 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 |
+
| 1.504 | 1.0 | 17 | 1.5036 |
|
| 57 |
+
| 1.3344 | 2.0 | 34 | 1.3101 |
|
| 58 |
+
| 1.0895 | 3.0 | 51 | 1.1146 |
|
| 59 |
+
| 0.9755 | 4.0 | 68 | 1.0741 |
|
| 60 |
+
| 0.9637 | 5.0 | 85 | 1.0524 |
|
| 61 |
+
| 0.9215 | 6.0 | 102 | 1.0349 |
|
| 62 |
+
| 0.8984 | 7.0 | 119 | 1.0345 |
|
| 63 |
+
| 0.7983 | 8.0 | 136 | 1.0459 |
|
| 64 |
+
| 0.711 | 9.0 | 153 | 1.0750 |
|
| 65 |
+
| 0.6725 | 10.0 | 170 | 1.1344 |
|
| 66 |
+
| 0.629 | 11.0 | 187 | 1.1630 |
|
| 67 |
+
| 0.4573 | 12.0 | 204 | 1.2680 |
|
| 68 |
+
| 0.4754 | 13.0 | 221 | 1.2757 |
|
| 69 |
+
| 0.4236 | 14.0 | 238 | 1.3371 |
|
| 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 sha256:868af29cc3ad681a9585395d1b0383ec8222f7dc7174de2dac3208c0c18a866b
|
| 3 |
size 167832240
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"total_flos": 1.8224671283478528e+17,
|
| 4 |
+
"train_loss": 0.8704949083949337,
|
| 5 |
+
"train_runtime": 6511.5654,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 1.044,
|
| 8 |
+
"train_steps_per_second": 0.131
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 14.0,
|
| 3 |
+
"total_flos": 1.8224671283478528e+17,
|
| 4 |
+
"train_loss": 0.8704949083949337,
|
| 5 |
+
"train_runtime": 6511.5654,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 1.044,
|
| 8 |
+
"train_steps_per_second": 0.131
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1003 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.0345184803009033,
|
| 3 |
+
"best_model_checkpoint": "data/Llama-31-8B_task-2_180-samples_config-1_full/checkpoint-119",
|
| 4 |
+
"epoch": 14.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 238,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.058823529411764705,
|
| 13 |
+
"grad_norm": 0.5214345455169678,
|
| 14 |
+
"learning_rate": 1.1764705882352942e-06,
|
| 15 |
+
"loss": 1.5724,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11764705882352941,
|
| 20 |
+
"grad_norm": 0.5508020520210266,
|
| 21 |
+
"learning_rate": 2.3529411764705885e-06,
|
| 22 |
+
"loss": 1.6207,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.23529411764705882,
|
| 27 |
+
"grad_norm": 0.5937786102294922,
|
| 28 |
+
"learning_rate": 4.705882352941177e-06,
|
| 29 |
+
"loss": 1.6106,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.35294117647058826,
|
| 34 |
+
"grad_norm": 0.5013459324836731,
|
| 35 |
+
"learning_rate": 7.058823529411765e-06,
|
| 36 |
+
"loss": 1.6182,
|
| 37 |
+
"step": 6
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.47058823529411764,
|
| 41 |
+
"grad_norm": 0.4731301963329315,
|
| 42 |
+
"learning_rate": 9.411764705882354e-06,
|
| 43 |
+
"loss": 1.5894,
|
| 44 |
+
"step": 8
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5882352941176471,
|
| 48 |
+
"grad_norm": 0.4294460713863373,
|
| 49 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 50 |
+
"loss": 1.5504,
|
| 51 |
+
"step": 10
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.7058823529411765,
|
| 55 |
+
"grad_norm": 0.4912262260913849,
|
| 56 |
+
"learning_rate": 1.411764705882353e-05,
|
| 57 |
+
"loss": 1.5279,
|
| 58 |
+
"step": 12
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.8235294117647058,
|
| 62 |
+
"grad_norm": 0.5266876220703125,
|
| 63 |
+
"learning_rate": 1.647058823529412e-05,
|
| 64 |
+
"loss": 1.5391,
|
| 65 |
+
"step": 14
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.9411764705882353,
|
| 69 |
+
"grad_norm": 0.391021728515625,
|
| 70 |
+
"learning_rate": 1.8823529411764708e-05,
|
| 71 |
+
"loss": 1.504,
|
| 72 |
+
"step": 16
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"eval_loss": 1.5035704374313354,
|
| 77 |
+
"eval_runtime": 37.0347,
|
| 78 |
+
"eval_samples_per_second": 0.972,
|
| 79 |
+
"eval_steps_per_second": 0.972,
|
| 80 |
+
"step": 17
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.0588235294117647,
|
| 84 |
+
"grad_norm": 0.3921213746070862,
|
| 85 |
+
"learning_rate": 2.1176470588235296e-05,
|
| 86 |
+
"loss": 1.5014,
|
| 87 |
+
"step": 18
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.1764705882352942,
|
| 91 |
+
"grad_norm": 0.38555383682250977,
|
| 92 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 93 |
+
"loss": 1.4873,
|
| 94 |
+
"step": 20
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.2941176470588236,
|
| 98 |
+
"grad_norm": 0.2951497733592987,
|
| 99 |
+
"learning_rate": 2.5882352941176475e-05,
|
| 100 |
+
"loss": 1.4459,
|
| 101 |
+
"step": 22
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.4117647058823528,
|
| 105 |
+
"grad_norm": 0.31671950221061707,
|
| 106 |
+
"learning_rate": 2.823529411764706e-05,
|
| 107 |
+
"loss": 1.4021,
|
| 108 |
+
"step": 24
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.5294117647058822,
|
| 112 |
+
"grad_norm": 0.2338249385356903,
|
| 113 |
+
"learning_rate": 3.058823529411765e-05,
|
| 114 |
+
"loss": 1.3717,
|
| 115 |
+
"step": 26
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.6470588235294117,
|
| 119 |
+
"grad_norm": 0.2480524629354477,
|
| 120 |
+
"learning_rate": 3.294117647058824e-05,
|
| 121 |
+
"loss": 1.364,
|
| 122 |
+
"step": 28
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.7647058823529411,
|
| 126 |
+
"grad_norm": 0.29380786418914795,
|
| 127 |
+
"learning_rate": 3.529411764705883e-05,
|
| 128 |
+
"loss": 1.3607,
|
| 129 |
+
"step": 30
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.8823529411764706,
|
| 133 |
+
"grad_norm": 0.26109302043914795,
|
| 134 |
+
"learning_rate": 3.7647058823529415e-05,
|
| 135 |
+
"loss": 1.3166,
|
| 136 |
+
"step": 32
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 2.0,
|
| 140 |
+
"grad_norm": 0.2589414417743683,
|
| 141 |
+
"learning_rate": 4e-05,
|
| 142 |
+
"loss": 1.3344,
|
| 143 |
+
"step": 34
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.0,
|
| 147 |
+
"eval_loss": 1.3100626468658447,
|
| 148 |
+
"eval_runtime": 37.0282,
|
| 149 |
+
"eval_samples_per_second": 0.972,
|
| 150 |
+
"eval_steps_per_second": 0.972,
|
| 151 |
+
"step": 34
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.1176470588235294,
|
| 155 |
+
"grad_norm": 0.2652131915092468,
|
| 156 |
+
"learning_rate": 4.235294117647059e-05,
|
| 157 |
+
"loss": 1.2648,
|
| 158 |
+
"step": 36
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.235294117647059,
|
| 162 |
+
"grad_norm": 0.2305995672941208,
|
| 163 |
+
"learning_rate": 4.470588235294118e-05,
|
| 164 |
+
"loss": 1.2315,
|
| 165 |
+
"step": 38
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.3529411764705883,
|
| 169 |
+
"grad_norm": 0.22336632013320923,
|
| 170 |
+
"learning_rate": 4.705882352941177e-05,
|
| 171 |
+
"loss": 1.208,
|
| 172 |
+
"step": 40
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.4705882352941178,
|
| 176 |
+
"grad_norm": 0.25589871406555176,
|
| 177 |
+
"learning_rate": 4.9411764705882355e-05,
|
| 178 |
+
"loss": 1.1869,
|
| 179 |
+
"step": 42
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.588235294117647,
|
| 183 |
+
"grad_norm": 0.3676718771457672,
|
| 184 |
+
"learning_rate": 5.176470588235295e-05,
|
| 185 |
+
"loss": 1.1343,
|
| 186 |
+
"step": 44
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.7058823529411766,
|
| 190 |
+
"grad_norm": 0.308319628238678,
|
| 191 |
+
"learning_rate": 5.411764705882353e-05,
|
| 192 |
+
"loss": 1.1467,
|
| 193 |
+
"step": 46
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.8235294117647056,
|
| 197 |
+
"grad_norm": 0.2406923770904541,
|
| 198 |
+
"learning_rate": 5.647058823529412e-05,
|
| 199 |
+
"loss": 1.1131,
|
| 200 |
+
"step": 48
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.9411764705882355,
|
| 204 |
+
"grad_norm": 0.3291592299938202,
|
| 205 |
+
"learning_rate": 5.882352941176471e-05,
|
| 206 |
+
"loss": 1.0895,
|
| 207 |
+
"step": 50
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.0,
|
| 211 |
+
"eval_loss": 1.1146036386489868,
|
| 212 |
+
"eval_runtime": 36.9993,
|
| 213 |
+
"eval_samples_per_second": 0.973,
|
| 214 |
+
"eval_steps_per_second": 0.973,
|
| 215 |
+
"step": 51
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.0588235294117645,
|
| 219 |
+
"grad_norm": 0.16729487478733063,
|
| 220 |
+
"learning_rate": 6.11764705882353e-05,
|
| 221 |
+
"loss": 1.0574,
|
| 222 |
+
"step": 52
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.176470588235294,
|
| 226 |
+
"grad_norm": 0.15566912293434143,
|
| 227 |
+
"learning_rate": 6.352941176470588e-05,
|
| 228 |
+
"loss": 1.0398,
|
| 229 |
+
"step": 54
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 3.2941176470588234,
|
| 233 |
+
"grad_norm": 0.15833047032356262,
|
| 234 |
+
"learning_rate": 6.588235294117648e-05,
|
| 235 |
+
"loss": 1.0902,
|
| 236 |
+
"step": 56
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.411764705882353,
|
| 240 |
+
"grad_norm": 0.16289210319519043,
|
| 241 |
+
"learning_rate": 6.823529411764707e-05,
|
| 242 |
+
"loss": 1.0512,
|
| 243 |
+
"step": 58
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.5294117647058822,
|
| 247 |
+
"grad_norm": 0.16850221157073975,
|
| 248 |
+
"learning_rate": 7.058823529411765e-05,
|
| 249 |
+
"loss": 0.9916,
|
| 250 |
+
"step": 60
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 3.6470588235294117,
|
| 254 |
+
"grad_norm": 0.15705695748329163,
|
| 255 |
+
"learning_rate": 7.294117647058823e-05,
|
| 256 |
+
"loss": 1.0991,
|
| 257 |
+
"step": 62
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.764705882352941,
|
| 261 |
+
"grad_norm": 0.192249596118927,
|
| 262 |
+
"learning_rate": 7.529411764705883e-05,
|
| 263 |
+
"loss": 1.022,
|
| 264 |
+
"step": 64
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.8823529411764706,
|
| 268 |
+
"grad_norm": 0.18115530908107758,
|
| 269 |
+
"learning_rate": 7.764705882352942e-05,
|
| 270 |
+
"loss": 0.9873,
|
| 271 |
+
"step": 66
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 4.0,
|
| 275 |
+
"grad_norm": 0.17344814538955688,
|
| 276 |
+
"learning_rate": 8e-05,
|
| 277 |
+
"loss": 0.9755,
|
| 278 |
+
"step": 68
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 4.0,
|
| 282 |
+
"eval_loss": 1.0741065740585327,
|
| 283 |
+
"eval_runtime": 36.9986,
|
| 284 |
+
"eval_samples_per_second": 0.973,
|
| 285 |
+
"eval_steps_per_second": 0.973,
|
| 286 |
+
"step": 68
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 4.117647058823529,
|
| 290 |
+
"grad_norm": 0.16507934033870697,
|
| 291 |
+
"learning_rate": 8.23529411764706e-05,
|
| 292 |
+
"loss": 1.0732,
|
| 293 |
+
"step": 70
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 4.235294117647059,
|
| 297 |
+
"grad_norm": 0.17189058661460876,
|
| 298 |
+
"learning_rate": 8.470588235294118e-05,
|
| 299 |
+
"loss": 0.9364,
|
| 300 |
+
"step": 72
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.352941176470588,
|
| 304 |
+
"grad_norm": 0.2030569314956665,
|
| 305 |
+
"learning_rate": 8.705882352941177e-05,
|
| 306 |
+
"loss": 1.0275,
|
| 307 |
+
"step": 74
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 4.470588235294118,
|
| 311 |
+
"grad_norm": 0.1766645312309265,
|
| 312 |
+
"learning_rate": 8.941176470588236e-05,
|
| 313 |
+
"loss": 0.9621,
|
| 314 |
+
"step": 76
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.588235294117647,
|
| 318 |
+
"grad_norm": 0.19495844841003418,
|
| 319 |
+
"learning_rate": 9.176470588235295e-05,
|
| 320 |
+
"loss": 0.9856,
|
| 321 |
+
"step": 78
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.705882352941177,
|
| 325 |
+
"grad_norm": 0.17263540625572205,
|
| 326 |
+
"learning_rate": 9.411764705882353e-05,
|
| 327 |
+
"loss": 1.012,
|
| 328 |
+
"step": 80
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 4.823529411764706,
|
| 332 |
+
"grad_norm": 0.19147881865501404,
|
| 333 |
+
"learning_rate": 9.647058823529412e-05,
|
| 334 |
+
"loss": 0.9723,
|
| 335 |
+
"step": 82
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 4.9411764705882355,
|
| 339 |
+
"grad_norm": 0.19374307990074158,
|
| 340 |
+
"learning_rate": 9.882352941176471e-05,
|
| 341 |
+
"loss": 0.9637,
|
| 342 |
+
"step": 84
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 5.0,
|
| 346 |
+
"eval_loss": 1.052433967590332,
|
| 347 |
+
"eval_runtime": 37.0057,
|
| 348 |
+
"eval_samples_per_second": 0.973,
|
| 349 |
+
"eval_steps_per_second": 0.973,
|
| 350 |
+
"step": 85
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 5.0588235294117645,
|
| 354 |
+
"grad_norm": 0.2304454743862152,
|
| 355 |
+
"learning_rate": 9.99995783847866e-05,
|
| 356 |
+
"loss": 0.9125,
|
| 357 |
+
"step": 86
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 5.176470588235294,
|
| 361 |
+
"grad_norm": 0.2263410985469818,
|
| 362 |
+
"learning_rate": 9.999620550574153e-05,
|
| 363 |
+
"loss": 0.942,
|
| 364 |
+
"step": 88
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 5.294117647058823,
|
| 368 |
+
"grad_norm": 0.2422245293855667,
|
| 369 |
+
"learning_rate": 9.998945997517956e-05,
|
| 370 |
+
"loss": 0.9778,
|
| 371 |
+
"step": 90
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 5.411764705882353,
|
| 375 |
+
"grad_norm": 0.20600196719169617,
|
| 376 |
+
"learning_rate": 9.997934224814173e-05,
|
| 377 |
+
"loss": 0.9842,
|
| 378 |
+
"step": 92
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.529411764705882,
|
| 382 |
+
"grad_norm": 0.23495109379291534,
|
| 383 |
+
"learning_rate": 9.996585300715116e-05,
|
| 384 |
+
"loss": 0.9389,
|
| 385 |
+
"step": 94
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 5.647058823529412,
|
| 389 |
+
"grad_norm": 0.27190443873405457,
|
| 390 |
+
"learning_rate": 9.994899316216708e-05,
|
| 391 |
+
"loss": 0.9416,
|
| 392 |
+
"step": 96
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 5.764705882352941,
|
| 396 |
+
"grad_norm": 0.24846382439136505,
|
| 397 |
+
"learning_rate": 9.992876385052345e-05,
|
| 398 |
+
"loss": 0.9654,
|
| 399 |
+
"step": 98
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 5.882352941176471,
|
| 403 |
+
"grad_norm": 0.23965619504451752,
|
| 404 |
+
"learning_rate": 9.990516643685222e-05,
|
| 405 |
+
"loss": 0.9653,
|
| 406 |
+
"step": 100
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 6.0,
|
| 410 |
+
"grad_norm": 0.23243875801563263,
|
| 411 |
+
"learning_rate": 9.987820251299122e-05,
|
| 412 |
+
"loss": 0.9215,
|
| 413 |
+
"step": 102
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.0,
|
| 417 |
+
"eval_loss": 1.0348577499389648,
|
| 418 |
+
"eval_runtime": 36.9988,
|
| 419 |
+
"eval_samples_per_second": 0.973,
|
| 420 |
+
"eval_steps_per_second": 0.973,
|
| 421 |
+
"step": 102
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 6.117647058823529,
|
| 425 |
+
"grad_norm": 0.2187218964099884,
|
| 426 |
+
"learning_rate": 9.984787389787688e-05,
|
| 427 |
+
"loss": 0.886,
|
| 428 |
+
"step": 104
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.235294117647059,
|
| 432 |
+
"grad_norm": 0.24537856876850128,
|
| 433 |
+
"learning_rate": 9.981418263742148e-05,
|
| 434 |
+
"loss": 0.8928,
|
| 435 |
+
"step": 106
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 6.352941176470588,
|
| 439 |
+
"grad_norm": 0.25448307394981384,
|
| 440 |
+
"learning_rate": 9.977713100437509e-05,
|
| 441 |
+
"loss": 0.8498,
|
| 442 |
+
"step": 108
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 6.470588235294118,
|
| 446 |
+
"grad_norm": 0.2609221339225769,
|
| 447 |
+
"learning_rate": 9.973672149817232e-05,
|
| 448 |
+
"loss": 0.926,
|
| 449 |
+
"step": 110
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 6.588235294117647,
|
| 453 |
+
"grad_norm": 0.27997028827667236,
|
| 454 |
+
"learning_rate": 9.96929568447637e-05,
|
| 455 |
+
"loss": 0.8778,
|
| 456 |
+
"step": 112
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 6.705882352941177,
|
| 460 |
+
"grad_norm": 0.29801490902900696,
|
| 461 |
+
"learning_rate": 9.964583999643174e-05,
|
| 462 |
+
"loss": 0.9633,
|
| 463 |
+
"step": 114
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 6.823529411764706,
|
| 467 |
+
"grad_norm": 0.25596335530281067,
|
| 468 |
+
"learning_rate": 9.95953741315919e-05,
|
| 469 |
+
"loss": 0.9494,
|
| 470 |
+
"step": 116
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 6.9411764705882355,
|
| 474 |
+
"grad_norm": 0.2853046655654907,
|
| 475 |
+
"learning_rate": 9.954156265457801e-05,
|
| 476 |
+
"loss": 0.8984,
|
| 477 |
+
"step": 118
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 7.0,
|
| 481 |
+
"eval_loss": 1.0345184803009033,
|
| 482 |
+
"eval_runtime": 36.9963,
|
| 483 |
+
"eval_samples_per_second": 0.973,
|
| 484 |
+
"eval_steps_per_second": 0.973,
|
| 485 |
+
"step": 119
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 7.0588235294117645,
|
| 489 |
+
"grad_norm": 0.31706833839416504,
|
| 490 |
+
"learning_rate": 9.948440919541278e-05,
|
| 491 |
+
"loss": 0.8992,
|
| 492 |
+
"step": 120
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.176470588235294,
|
| 496 |
+
"grad_norm": 0.2826850712299347,
|
| 497 |
+
"learning_rate": 9.942391760956277e-05,
|
| 498 |
+
"loss": 0.8611,
|
| 499 |
+
"step": 122
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 7.294117647058823,
|
| 503 |
+
"grad_norm": 0.33343079686164856,
|
| 504 |
+
"learning_rate": 9.936009197767845e-05,
|
| 505 |
+
"loss": 0.8488,
|
| 506 |
+
"step": 124
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.411764705882353,
|
| 510 |
+
"grad_norm": 0.33269575238227844,
|
| 511 |
+
"learning_rate": 9.929293660531888e-05,
|
| 512 |
+
"loss": 0.809,
|
| 513 |
+
"step": 126
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.529411764705882,
|
| 517 |
+
"grad_norm": 0.3902694582939148,
|
| 518 |
+
"learning_rate": 9.922245602266118e-05,
|
| 519 |
+
"loss": 0.8352,
|
| 520 |
+
"step": 128
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.647058823529412,
|
| 524 |
+
"grad_norm": 0.34491705894470215,
|
| 525 |
+
"learning_rate": 9.91486549841951e-05,
|
| 526 |
+
"loss": 0.8561,
|
| 527 |
+
"step": 130
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.764705882352941,
|
| 531 |
+
"grad_norm": 0.3838919401168823,
|
| 532 |
+
"learning_rate": 9.90715384684021e-05,
|
| 533 |
+
"loss": 0.7896,
|
| 534 |
+
"step": 132
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.882352941176471,
|
| 538 |
+
"grad_norm": 0.3754192590713501,
|
| 539 |
+
"learning_rate": 9.899111167741966e-05,
|
| 540 |
+
"loss": 0.9033,
|
| 541 |
+
"step": 134
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 8.0,
|
| 545 |
+
"grad_norm": 0.3832869529724121,
|
| 546 |
+
"learning_rate": 9.890738003669029e-05,
|
| 547 |
+
"loss": 0.7983,
|
| 548 |
+
"step": 136
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 8.0,
|
| 552 |
+
"eval_loss": 1.0459100008010864,
|
| 553 |
+
"eval_runtime": 37.0077,
|
| 554 |
+
"eval_samples_per_second": 0.973,
|
| 555 |
+
"eval_steps_per_second": 0.973,
|
| 556 |
+
"step": 136
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 8.117647058823529,
|
| 560 |
+
"grad_norm": 0.4289301633834839,
|
| 561 |
+
"learning_rate": 9.882034919459555e-05,
|
| 562 |
+
"loss": 0.7896,
|
| 563 |
+
"step": 138
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 8.235294117647058,
|
| 567 |
+
"grad_norm": 0.5225833058357239,
|
| 568 |
+
"learning_rate": 9.873002502207503e-05,
|
| 569 |
+
"loss": 0.7458,
|
| 570 |
+
"step": 140
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 8.352941176470589,
|
| 574 |
+
"grad_norm": 0.47407665848731995,
|
| 575 |
+
"learning_rate": 9.863641361223024e-05,
|
| 576 |
+
"loss": 0.8066,
|
| 577 |
+
"step": 142
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 8.470588235294118,
|
| 581 |
+
"grad_norm": 0.46604663133621216,
|
| 582 |
+
"learning_rate": 9.853952127991372e-05,
|
| 583 |
+
"loss": 0.7907,
|
| 584 |
+
"step": 144
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 8.588235294117647,
|
| 588 |
+
"grad_norm": 0.4920913577079773,
|
| 589 |
+
"learning_rate": 9.843935456130295e-05,
|
| 590 |
+
"loss": 0.8002,
|
| 591 |
+
"step": 146
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 8.705882352941176,
|
| 595 |
+
"grad_norm": 0.4699203670024872,
|
| 596 |
+
"learning_rate": 9.833592021345937e-05,
|
| 597 |
+
"loss": 0.7876,
|
| 598 |
+
"step": 148
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 8.823529411764707,
|
| 602 |
+
"grad_norm": 0.4708077907562256,
|
| 603 |
+
"learning_rate": 9.822922521387276e-05,
|
| 604 |
+
"loss": 0.7274,
|
| 605 |
+
"step": 150
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 8.941176470588236,
|
| 609 |
+
"grad_norm": 0.6590410470962524,
|
| 610 |
+
"learning_rate": 9.811927675999036e-05,
|
| 611 |
+
"loss": 0.711,
|
| 612 |
+
"step": 152
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 9.0,
|
| 616 |
+
"eval_loss": 1.0750151872634888,
|
| 617 |
+
"eval_runtime": 37.003,
|
| 618 |
+
"eval_samples_per_second": 0.973,
|
| 619 |
+
"eval_steps_per_second": 0.973,
|
| 620 |
+
"step": 153
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 9.058823529411764,
|
| 624 |
+
"grad_norm": 0.6144815683364868,
|
| 625 |
+
"learning_rate": 9.800608226873142e-05,
|
| 626 |
+
"loss": 0.7504,
|
| 627 |
+
"step": 154
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 9.176470588235293,
|
| 631 |
+
"grad_norm": 0.5911457538604736,
|
| 632 |
+
"learning_rate": 9.788964937598689e-05,
|
| 633 |
+
"loss": 0.6873,
|
| 634 |
+
"step": 156
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 9.294117647058824,
|
| 638 |
+
"grad_norm": 0.6449252963066101,
|
| 639 |
+
"learning_rate": 9.776998593610428e-05,
|
| 640 |
+
"loss": 0.7058,
|
| 641 |
+
"step": 158
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 9.411764705882353,
|
| 645 |
+
"grad_norm": 0.5975269079208374,
|
| 646 |
+
"learning_rate": 9.764710002135784e-05,
|
| 647 |
+
"loss": 0.7004,
|
| 648 |
+
"step": 160
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 9.529411764705882,
|
| 652 |
+
"grad_norm": 0.5621377825737,
|
| 653 |
+
"learning_rate": 9.752099992140399e-05,
|
| 654 |
+
"loss": 0.672,
|
| 655 |
+
"step": 162
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 9.647058823529411,
|
| 659 |
+
"grad_norm": 0.6631340980529785,
|
| 660 |
+
"learning_rate": 9.739169414272217e-05,
|
| 661 |
+
"loss": 0.6917,
|
| 662 |
+
"step": 164
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 9.764705882352942,
|
| 666 |
+
"grad_norm": 0.6114043593406677,
|
| 667 |
+
"learning_rate": 9.725919140804099e-05,
|
| 668 |
+
"loss": 0.7236,
|
| 669 |
+
"step": 166
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 9.882352941176471,
|
| 673 |
+
"grad_norm": 0.6280692219734192,
|
| 674 |
+
"learning_rate": 9.71235006557497e-05,
|
| 675 |
+
"loss": 0.602,
|
| 676 |
+
"step": 168
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 10.0,
|
| 680 |
+
"grad_norm": 0.6364344358444214,
|
| 681 |
+
"learning_rate": 9.698463103929542e-05,
|
| 682 |
+
"loss": 0.6725,
|
| 683 |
+
"step": 170
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 10.0,
|
| 687 |
+
"eval_loss": 1.134390950202942,
|
| 688 |
+
"eval_runtime": 37.0,
|
| 689 |
+
"eval_samples_per_second": 0.973,
|
| 690 |
+
"eval_steps_per_second": 0.973,
|
| 691 |
+
"step": 170
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 10.117647058823529,
|
| 695 |
+
"grad_norm": 0.6386916041374207,
|
| 696 |
+
"learning_rate": 9.684259192656553e-05,
|
| 697 |
+
"loss": 0.561,
|
| 698 |
+
"step": 172
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 10.235294117647058,
|
| 702 |
+
"grad_norm": 0.9274471402168274,
|
| 703 |
+
"learning_rate": 9.669739289925577e-05,
|
| 704 |
+
"loss": 0.5975,
|
| 705 |
+
"step": 174
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 10.352941176470589,
|
| 709 |
+
"grad_norm": 0.758702278137207,
|
| 710 |
+
"learning_rate": 9.654904375222385e-05,
|
| 711 |
+
"loss": 0.621,
|
| 712 |
+
"step": 176
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 10.470588235294118,
|
| 716 |
+
"grad_norm": 0.949181854724884,
|
| 717 |
+
"learning_rate": 9.639755449282875e-05,
|
| 718 |
+
"loss": 0.5464,
|
| 719 |
+
"step": 178
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 10.588235294117647,
|
| 723 |
+
"grad_norm": 0.7223683595657349,
|
| 724 |
+
"learning_rate": 9.62429353402556e-05,
|
| 725 |
+
"loss": 0.6089,
|
| 726 |
+
"step": 180
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 10.705882352941176,
|
| 730 |
+
"grad_norm": 0.6299746632575989,
|
| 731 |
+
"learning_rate": 9.608519672482636e-05,
|
| 732 |
+
"loss": 0.5711,
|
| 733 |
+
"step": 182
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 10.823529411764707,
|
| 737 |
+
"grad_norm": 0.7116110920906067,
|
| 738 |
+
"learning_rate": 9.592434928729616e-05,
|
| 739 |
+
"loss": 0.5963,
|
| 740 |
+
"step": 184
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 10.941176470588236,
|
| 744 |
+
"grad_norm": 0.6831782460212708,
|
| 745 |
+
"learning_rate": 9.576040387813552e-05,
|
| 746 |
+
"loss": 0.629,
|
| 747 |
+
"step": 186
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 11.0,
|
| 751 |
+
"eval_loss": 1.1629810333251953,
|
| 752 |
+
"eval_runtime": 37.0044,
|
| 753 |
+
"eval_samples_per_second": 0.973,
|
| 754 |
+
"eval_steps_per_second": 0.973,
|
| 755 |
+
"step": 187
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 11.058823529411764,
|
| 759 |
+
"grad_norm": 0.6914322972297668,
|
| 760 |
+
"learning_rate": 9.559337155679842e-05,
|
| 761 |
+
"loss": 0.6061,
|
| 762 |
+
"step": 188
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 11.176470588235293,
|
| 766 |
+
"grad_norm": 0.8476060628890991,
|
| 767 |
+
"learning_rate": 9.542326359097619e-05,
|
| 768 |
+
"loss": 0.513,
|
| 769 |
+
"step": 190
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 11.294117647058824,
|
| 773 |
+
"grad_norm": 0.9703707098960876,
|
| 774 |
+
"learning_rate": 9.525009145583745e-05,
|
| 775 |
+
"loss": 0.5803,
|
| 776 |
+
"step": 192
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 11.411764705882353,
|
| 780 |
+
"grad_norm": 0.8944510221481323,
|
| 781 |
+
"learning_rate": 9.507386683325404e-05,
|
| 782 |
+
"loss": 0.3622,
|
| 783 |
+
"step": 194
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 11.529411764705882,
|
| 787 |
+
"grad_norm": 1.0056551694869995,
|
| 788 |
+
"learning_rate": 9.489460161101291e-05,
|
| 789 |
+
"loss": 0.5537,
|
| 790 |
+
"step": 196
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 11.647058823529411,
|
| 794 |
+
"grad_norm": 0.9109512567520142,
|
| 795 |
+
"learning_rate": 9.471230788201429e-05,
|
| 796 |
+
"loss": 0.519,
|
| 797 |
+
"step": 198
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 11.764705882352942,
|
| 801 |
+
"grad_norm": 0.82012540102005,
|
| 802 |
+
"learning_rate": 9.452699794345581e-05,
|
| 803 |
+
"loss": 0.5082,
|
| 804 |
+
"step": 200
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 11.882352941176471,
|
| 808 |
+
"grad_norm": 0.7492274641990662,
|
| 809 |
+
"learning_rate": 9.43386842960031e-05,
|
| 810 |
+
"loss": 0.5116,
|
| 811 |
+
"step": 202
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 12.0,
|
| 815 |
+
"grad_norm": 0.8394200801849365,
|
| 816 |
+
"learning_rate": 9.414737964294636e-05,
|
| 817 |
+
"loss": 0.4573,
|
| 818 |
+
"step": 204
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 12.0,
|
| 822 |
+
"eval_loss": 1.2680054903030396,
|
| 823 |
+
"eval_runtime": 37.0068,
|
| 824 |
+
"eval_samples_per_second": 0.973,
|
| 825 |
+
"eval_steps_per_second": 0.973,
|
| 826 |
+
"step": 204
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 12.117647058823529,
|
| 830 |
+
"grad_norm": 0.8289351463317871,
|
| 831 |
+
"learning_rate": 9.395309688934351e-05,
|
| 832 |
+
"loss": 0.4454,
|
| 833 |
+
"step": 206
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 12.235294117647058,
|
| 837 |
+
"grad_norm": 0.97170090675354,
|
| 838 |
+
"learning_rate": 9.375584914114963e-05,
|
| 839 |
+
"loss": 0.3926,
|
| 840 |
+
"step": 208
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 12.352941176470589,
|
| 844 |
+
"grad_norm": 0.8402796983718872,
|
| 845 |
+
"learning_rate": 9.355564970433288e-05,
|
| 846 |
+
"loss": 0.4302,
|
| 847 |
+
"step": 210
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 12.470588235294118,
|
| 851 |
+
"grad_norm": 0.8999544978141785,
|
| 852 |
+
"learning_rate": 9.335251208397684e-05,
|
| 853 |
+
"loss": 0.3877,
|
| 854 |
+
"step": 212
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 12.588235294117647,
|
| 858 |
+
"grad_norm": 1.080120325088501,
|
| 859 |
+
"learning_rate": 9.314644998336949e-05,
|
| 860 |
+
"loss": 0.4688,
|
| 861 |
+
"step": 214
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 12.705882352941176,
|
| 865 |
+
"grad_norm": 0.8625339269638062,
|
| 866 |
+
"learning_rate": 9.293747730307889e-05,
|
| 867 |
+
"loss": 0.3901,
|
| 868 |
+
"step": 216
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 12.823529411764707,
|
| 872 |
+
"grad_norm": 1.0037257671356201,
|
| 873 |
+
"learning_rate": 9.272560814001539e-05,
|
| 874 |
+
"loss": 0.4283,
|
| 875 |
+
"step": 218
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 12.941176470588236,
|
| 879 |
+
"grad_norm": 0.8916875720024109,
|
| 880 |
+
"learning_rate": 9.251085678648072e-05,
|
| 881 |
+
"loss": 0.4754,
|
| 882 |
+
"step": 220
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 13.0,
|
| 886 |
+
"eval_loss": 1.2756708860397339,
|
| 887 |
+
"eval_runtime": 37.001,
|
| 888 |
+
"eval_samples_per_second": 0.973,
|
| 889 |
+
"eval_steps_per_second": 0.973,
|
| 890 |
+
"step": 221
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 13.058823529411764,
|
| 894 |
+
"grad_norm": 0.9820285439491272,
|
| 895 |
+
"learning_rate": 9.229323772920381e-05,
|
| 896 |
+
"loss": 0.4135,
|
| 897 |
+
"step": 222
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 13.176470588235293,
|
| 901 |
+
"grad_norm": 1.5194408893585205,
|
| 902 |
+
"learning_rate": 9.207276564836366e-05,
|
| 903 |
+
"loss": 0.3795,
|
| 904 |
+
"step": 224
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 13.294117647058824,
|
| 908 |
+
"grad_norm": 0.8674440383911133,
|
| 909 |
+
"learning_rate": 9.184945541659889e-05,
|
| 910 |
+
"loss": 0.3577,
|
| 911 |
+
"step": 226
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 13.411764705882353,
|
| 915 |
+
"grad_norm": 1.037719488143921,
|
| 916 |
+
"learning_rate": 9.162332209800455e-05,
|
| 917 |
+
"loss": 0.292,
|
| 918 |
+
"step": 228
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 13.529411764705882,
|
| 922 |
+
"grad_norm": 1.3491851091384888,
|
| 923 |
+
"learning_rate": 9.139438094711589e-05,
|
| 924 |
+
"loss": 0.3822,
|
| 925 |
+
"step": 230
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 13.647058823529411,
|
| 929 |
+
"grad_norm": 1.114529013633728,
|
| 930 |
+
"learning_rate": 9.116264740787936e-05,
|
| 931 |
+
"loss": 0.3377,
|
| 932 |
+
"step": 232
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 13.764705882352942,
|
| 936 |
+
"grad_norm": 0.971333920955658,
|
| 937 |
+
"learning_rate": 9.092813711261074e-05,
|
| 938 |
+
"loss": 0.4037,
|
| 939 |
+
"step": 234
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 13.882352941176471,
|
| 943 |
+
"grad_norm": 1.0582764148712158,
|
| 944 |
+
"learning_rate": 9.069086588094067e-05,
|
| 945 |
+
"loss": 0.3383,
|
| 946 |
+
"step": 236
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 14.0,
|
| 950 |
+
"grad_norm": 0.9523320198059082,
|
| 951 |
+
"learning_rate": 9.045084971874738e-05,
|
| 952 |
+
"loss": 0.4236,
|
| 953 |
+
"step": 238
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 14.0,
|
| 957 |
+
"eval_loss": 1.337109923362732,
|
| 958 |
+
"eval_runtime": 37.033,
|
| 959 |
+
"eval_samples_per_second": 0.972,
|
| 960 |
+
"eval_steps_per_second": 0.972,
|
| 961 |
+
"step": 238
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 14.0,
|
| 965 |
+
"step": 238,
|
| 966 |
+
"total_flos": 1.8224671283478528e+17,
|
| 967 |
+
"train_loss": 0.8704949083949337,
|
| 968 |
+
"train_runtime": 6511.5654,
|
| 969 |
+
"train_samples_per_second": 1.044,
|
| 970 |
+
"train_steps_per_second": 0.131
|
| 971 |
+
}
|
| 972 |
+
],
|
| 973 |
+
"logging_steps": 2,
|
| 974 |
+
"max_steps": 850,
|
| 975 |
+
"num_input_tokens_seen": 0,
|
| 976 |
+
"num_train_epochs": 50,
|
| 977 |
+
"save_steps": 25,
|
| 978 |
+
"stateful_callbacks": {
|
| 979 |
+
"EarlyStoppingCallback": {
|
| 980 |
+
"args": {
|
| 981 |
+
"early_stopping_patience": 7,
|
| 982 |
+
"early_stopping_threshold": 0.0
|
| 983 |
+
},
|
| 984 |
+
"attributes": {
|
| 985 |
+
"early_stopping_patience_counter": 0
|
| 986 |
+
}
|
| 987 |
+
},
|
| 988 |
+
"TrainerControl": {
|
| 989 |
+
"args": {
|
| 990 |
+
"should_epoch_stop": false,
|
| 991 |
+
"should_evaluate": false,
|
| 992 |
+
"should_log": false,
|
| 993 |
+
"should_save": true,
|
| 994 |
+
"should_training_stop": true
|
| 995 |
+
},
|
| 996 |
+
"attributes": {}
|
| 997 |
+
}
|
| 998 |
+
},
|
| 999 |
+
"total_flos": 1.8224671283478528e+17,
|
| 1000 |
+
"train_batch_size": 1,
|
| 1001 |
+
"trial_name": null,
|
| 1002 |
+
"trial_params": null
|
| 1003 |
+
}
|