Commit
•
6253f58
1
Parent(s):
5d34a90
Model save
Browse files- README.md +90 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1813 -0
README.md
ADDED
|
@@ -0,0 +1,90 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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-3
|
| 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-3
|
| 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: 0.8289
|
| 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: 1e-05
|
| 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: 150
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 56 |
+
| 1.0365 | 1.0 | 17 | 1.1316 |
|
| 57 |
+
| 1.1746 | 2.0 | 34 | 1.1196 |
|
| 58 |
+
| 1.0933 | 3.0 | 51 | 1.0957 |
|
| 59 |
+
| 0.985 | 4.0 | 68 | 1.0540 |
|
| 60 |
+
| 0.9741 | 5.0 | 85 | 0.9950 |
|
| 61 |
+
| 1.0008 | 6.0 | 102 | 0.9377 |
|
| 62 |
+
| 0.8935 | 7.0 | 119 | 0.8939 |
|
| 63 |
+
| 0.8862 | 8.0 | 136 | 0.8579 |
|
| 64 |
+
| 0.8266 | 9.0 | 153 | 0.8294 |
|
| 65 |
+
| 0.7797 | 10.0 | 170 | 0.8075 |
|
| 66 |
+
| 0.8158 | 11.0 | 187 | 0.7903 |
|
| 67 |
+
| 0.6845 | 12.0 | 204 | 0.7742 |
|
| 68 |
+
| 0.6819 | 13.0 | 221 | 0.7598 |
|
| 69 |
+
| 0.7241 | 14.0 | 238 | 0.7472 |
|
| 70 |
+
| 0.695 | 15.0 | 255 | 0.7365 |
|
| 71 |
+
| 0.6982 | 16.0 | 272 | 0.7272 |
|
| 72 |
+
| 0.622 | 17.0 | 289 | 0.7215 |
|
| 73 |
+
| 0.5905 | 18.0 | 306 | 0.7156 |
|
| 74 |
+
| 0.6121 | 19.0 | 323 | 0.7140 |
|
| 75 |
+
| 0.567 | 20.0 | 340 | 0.7166 |
|
| 76 |
+
| 0.5471 | 21.0 | 357 | 0.7172 |
|
| 77 |
+
| 0.4761 | 22.0 | 374 | 0.7234 |
|
| 78 |
+
| 0.4967 | 23.0 | 391 | 0.7358 |
|
| 79 |
+
| 0.4833 | 24.0 | 408 | 0.7644 |
|
| 80 |
+
| 0.4071 | 25.0 | 425 | 0.8012 |
|
| 81 |
+
| 0.3567 | 26.0 | 442 | 0.8289 |
|
| 82 |
+
|
| 83 |
+
|
| 84 |
+
### Framework versions
|
| 85 |
+
|
| 86 |
+
- PEFT 0.12.0
|
| 87 |
+
- Transformers 4.44.0
|
| 88 |
+
- Pytorch 2.1.2+cu121
|
| 89 |
+
- Datasets 2.20.0
|
| 90 |
+
- 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 sha256:417ecf2e864cb4dd0073d3d5bb288f9a5f59f201cbb62fc4f0134fa633622d66
|
| 3 |
size 167832240
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 26.0,
|
| 3 |
+
"total_flos": 3.38458179962667e+17,
|
| 4 |
+
"train_loss": 0.7497456199173475,
|
| 5 |
+
"train_runtime": 12081.7026,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 1.689,
|
| 8 |
+
"train_steps_per_second": 0.211
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 26.0,
|
| 3 |
+
"total_flos": 3.38458179962667e+17,
|
| 4 |
+
"train_loss": 0.7497456199173475,
|
| 5 |
+
"train_runtime": 12081.7026,
|
| 6 |
+
"train_samples": 136,
|
| 7 |
+
"train_samples_per_second": 1.689,
|
| 8 |
+
"train_steps_per_second": 0.211
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1813 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7140286564826965,
|
| 3 |
+
"best_model_checkpoint": "data/Llama-31-8B_task-2_180-samples_config-3/checkpoint-323",
|
| 4 |
+
"epoch": 26.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 442,
|
| 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": 0.7734732031822205,
|
| 14 |
+
"learning_rate": 3.9215686274509804e-08,
|
| 15 |
+
"loss": 1.1412,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.11764705882352941,
|
| 20 |
+
"grad_norm": 0.6782335638999939,
|
| 21 |
+
"learning_rate": 7.843137254901961e-08,
|
| 22 |
+
"loss": 1.1733,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.23529411764705882,
|
| 27 |
+
"grad_norm": 0.8408952951431274,
|
| 28 |
+
"learning_rate": 1.5686274509803921e-07,
|
| 29 |
+
"loss": 1.1666,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.35294117647058826,
|
| 34 |
+
"grad_norm": 0.7287867069244385,
|
| 35 |
+
"learning_rate": 2.3529411764705883e-07,
|
| 36 |
+
"loss": 1.1801,
|
| 37 |
+
"step": 6
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.47058823529411764,
|
| 41 |
+
"grad_norm": 0.7565770745277405,
|
| 42 |
+
"learning_rate": 3.1372549019607843e-07,
|
| 43 |
+
"loss": 1.1116,
|
| 44 |
+
"step": 8
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5882352941176471,
|
| 48 |
+
"grad_norm": 0.6836811304092407,
|
| 49 |
+
"learning_rate": 3.921568627450981e-07,
|
| 50 |
+
"loss": 1.1475,
|
| 51 |
+
"step": 10
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.7058823529411765,
|
| 55 |
+
"grad_norm": 0.6788910627365112,
|
| 56 |
+
"learning_rate": 4.7058823529411767e-07,
|
| 57 |
+
"loss": 1.0752,
|
| 58 |
+
"step": 12
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.8235294117647058,
|
| 62 |
+
"grad_norm": 0.6463932991027832,
|
| 63 |
+
"learning_rate": 5.490196078431373e-07,
|
| 64 |
+
"loss": 1.1428,
|
| 65 |
+
"step": 14
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.9411764705882353,
|
| 69 |
+
"grad_norm": 0.7503110766410828,
|
| 70 |
+
"learning_rate": 6.274509803921569e-07,
|
| 71 |
+
"loss": 1.0365,
|
| 72 |
+
"step": 16
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"eval_loss": 1.131649374961853,
|
| 77 |
+
"eval_runtime": 37.0173,
|
| 78 |
+
"eval_samples_per_second": 0.973,
|
| 79 |
+
"eval_steps_per_second": 0.973,
|
| 80 |
+
"step": 17
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.0588235294117647,
|
| 84 |
+
"grad_norm": 0.7130016088485718,
|
| 85 |
+
"learning_rate": 7.058823529411766e-07,
|
| 86 |
+
"loss": 1.1439,
|
| 87 |
+
"step": 18
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.1764705882352942,
|
| 91 |
+
"grad_norm": 0.69338458776474,
|
| 92 |
+
"learning_rate": 7.843137254901962e-07,
|
| 93 |
+
"loss": 1.1374,
|
| 94 |
+
"step": 20
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.2941176470588236,
|
| 98 |
+
"grad_norm": 0.6517343521118164,
|
| 99 |
+
"learning_rate": 8.627450980392157e-07,
|
| 100 |
+
"loss": 1.1213,
|
| 101 |
+
"step": 22
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.4117647058823528,
|
| 105 |
+
"grad_norm": 0.685614287853241,
|
| 106 |
+
"learning_rate": 9.411764705882353e-07,
|
| 107 |
+
"loss": 1.1182,
|
| 108 |
+
"step": 24
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.5294117647058822,
|
| 112 |
+
"grad_norm": 0.6998293995857239,
|
| 113 |
+
"learning_rate": 1.019607843137255e-06,
|
| 114 |
+
"loss": 1.0508,
|
| 115 |
+
"step": 26
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.6470588235294117,
|
| 119 |
+
"grad_norm": 0.6757376194000244,
|
| 120 |
+
"learning_rate": 1.0980392156862745e-06,
|
| 121 |
+
"loss": 1.0773,
|
| 122 |
+
"step": 28
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.7647058823529411,
|
| 126 |
+
"grad_norm": 0.7003068327903748,
|
| 127 |
+
"learning_rate": 1.1764705882352942e-06,
|
| 128 |
+
"loss": 1.1765,
|
| 129 |
+
"step": 30
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.8823529411764706,
|
| 133 |
+
"grad_norm": 0.75652676820755,
|
| 134 |
+
"learning_rate": 1.2549019607843137e-06,
|
| 135 |
+
"loss": 1.0897,
|
| 136 |
+
"step": 32
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 2.0,
|
| 140 |
+
"grad_norm": 0.8910195231437683,
|
| 141 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 142 |
+
"loss": 1.1746,
|
| 143 |
+
"step": 34
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.0,
|
| 147 |
+
"eval_loss": 1.119631290435791,
|
| 148 |
+
"eval_runtime": 37.0058,
|
| 149 |
+
"eval_samples_per_second": 0.973,
|
| 150 |
+
"eval_steps_per_second": 0.973,
|
| 151 |
+
"step": 34
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.1176470588235294,
|
| 155 |
+
"grad_norm": 0.6963354349136353,
|
| 156 |
+
"learning_rate": 1.4117647058823531e-06,
|
| 157 |
+
"loss": 1.0914,
|
| 158 |
+
"step": 36
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.235294117647059,
|
| 162 |
+
"grad_norm": 0.6639174818992615,
|
| 163 |
+
"learning_rate": 1.4901960784313726e-06,
|
| 164 |
+
"loss": 1.1038,
|
| 165 |
+
"step": 38
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.3529411764705883,
|
| 169 |
+
"grad_norm": 0.7396222949028015,
|
| 170 |
+
"learning_rate": 1.5686274509803923e-06,
|
| 171 |
+
"loss": 1.0596,
|
| 172 |
+
"step": 40
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.4705882352941178,
|
| 176 |
+
"grad_norm": 0.7213342189788818,
|
| 177 |
+
"learning_rate": 1.6470588235294118e-06,
|
| 178 |
+
"loss": 1.0962,
|
| 179 |
+
"step": 42
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.588235294117647,
|
| 183 |
+
"grad_norm": 0.7507779002189636,
|
| 184 |
+
"learning_rate": 1.7254901960784315e-06,
|
| 185 |
+
"loss": 1.0945,
|
| 186 |
+
"step": 44
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.7058823529411766,
|
| 190 |
+
"grad_norm": 0.803626298904419,
|
| 191 |
+
"learning_rate": 1.8039215686274512e-06,
|
| 192 |
+
"loss": 1.1417,
|
| 193 |
+
"step": 46
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.8235294117647056,
|
| 197 |
+
"grad_norm": 0.8166558742523193,
|
| 198 |
+
"learning_rate": 1.8823529411764707e-06,
|
| 199 |
+
"loss": 1.1391,
|
| 200 |
+
"step": 48
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.9411764705882355,
|
| 204 |
+
"grad_norm": 0.7504469752311707,
|
| 205 |
+
"learning_rate": 1.96078431372549e-06,
|
| 206 |
+
"loss": 1.0933,
|
| 207 |
+
"step": 50
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.0,
|
| 211 |
+
"eval_loss": 1.0957257747650146,
|
| 212 |
+
"eval_runtime": 37.0081,
|
| 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": 0.7341092228889465,
|
| 220 |
+
"learning_rate": 2.03921568627451e-06,
|
| 221 |
+
"loss": 1.0645,
|
| 222 |
+
"step": 52
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.176470588235294,
|
| 226 |
+
"grad_norm": 0.7262235283851624,
|
| 227 |
+
"learning_rate": 2.1176470588235296e-06,
|
| 228 |
+
"loss": 1.1047,
|
| 229 |
+
"step": 54
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 3.2941176470588234,
|
| 233 |
+
"grad_norm": 0.6933767795562744,
|
| 234 |
+
"learning_rate": 2.196078431372549e-06,
|
| 235 |
+
"loss": 1.0305,
|
| 236 |
+
"step": 56
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.411764705882353,
|
| 240 |
+
"grad_norm": 0.8481059670448303,
|
| 241 |
+
"learning_rate": 2.274509803921569e-06,
|
| 242 |
+
"loss": 1.1403,
|
| 243 |
+
"step": 58
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.5294117647058822,
|
| 247 |
+
"grad_norm": 0.7597863078117371,
|
| 248 |
+
"learning_rate": 2.3529411764705885e-06,
|
| 249 |
+
"loss": 1.0537,
|
| 250 |
+
"step": 60
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 3.6470588235294117,
|
| 254 |
+
"grad_norm": 0.7542691826820374,
|
| 255 |
+
"learning_rate": 2.431372549019608e-06,
|
| 256 |
+
"loss": 1.0815,
|
| 257 |
+
"step": 62
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.764705882352941,
|
| 261 |
+
"grad_norm": 0.7457835674285889,
|
| 262 |
+
"learning_rate": 2.5098039215686274e-06,
|
| 263 |
+
"loss": 1.0747,
|
| 264 |
+
"step": 64
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.8823529411764706,
|
| 268 |
+
"grad_norm": 0.6987655758857727,
|
| 269 |
+
"learning_rate": 2.5882352941176473e-06,
|
| 270 |
+
"loss": 1.1015,
|
| 271 |
+
"step": 66
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 4.0,
|
| 275 |
+
"grad_norm": 0.7653221487998962,
|
| 276 |
+
"learning_rate": 2.666666666666667e-06,
|
| 277 |
+
"loss": 0.985,
|
| 278 |
+
"step": 68
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 4.0,
|
| 282 |
+
"eval_loss": 1.0540263652801514,
|
| 283 |
+
"eval_runtime": 37.0036,
|
| 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": 0.6138588190078735,
|
| 291 |
+
"learning_rate": 2.7450980392156867e-06,
|
| 292 |
+
"loss": 1.0818,
|
| 293 |
+
"step": 70
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 4.235294117647059,
|
| 297 |
+
"grad_norm": 0.7161767482757568,
|
| 298 |
+
"learning_rate": 2.8235294117647062e-06,
|
| 299 |
+
"loss": 1.0233,
|
| 300 |
+
"step": 72
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.352941176470588,
|
| 304 |
+
"grad_norm": 0.5691388845443726,
|
| 305 |
+
"learning_rate": 2.901960784313726e-06,
|
| 306 |
+
"loss": 1.0152,
|
| 307 |
+
"step": 74
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 4.470588235294118,
|
| 311 |
+
"grad_norm": 0.6562293171882629,
|
| 312 |
+
"learning_rate": 2.980392156862745e-06,
|
| 313 |
+
"loss": 1.0299,
|
| 314 |
+
"step": 76
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.588235294117647,
|
| 318 |
+
"grad_norm": 0.6522495150566101,
|
| 319 |
+
"learning_rate": 3.058823529411765e-06,
|
| 320 |
+
"loss": 1.0341,
|
| 321 |
+
"step": 78
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.705882352941177,
|
| 325 |
+
"grad_norm": 0.6249567270278931,
|
| 326 |
+
"learning_rate": 3.1372549019607846e-06,
|
| 327 |
+
"loss": 0.9573,
|
| 328 |
+
"step": 80
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 4.823529411764706,
|
| 332 |
+
"grad_norm": 0.6513485312461853,
|
| 333 |
+
"learning_rate": 3.2156862745098045e-06,
|
| 334 |
+
"loss": 1.0373,
|
| 335 |
+
"step": 82
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 4.9411764705882355,
|
| 339 |
+
"grad_norm": 0.6425153613090515,
|
| 340 |
+
"learning_rate": 3.2941176470588236e-06,
|
| 341 |
+
"loss": 0.9741,
|
| 342 |
+
"step": 84
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 5.0,
|
| 346 |
+
"eval_loss": 0.99503093957901,
|
| 347 |
+
"eval_runtime": 37.0007,
|
| 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": 0.645844042301178,
|
| 355 |
+
"learning_rate": 3.3725490196078435e-06,
|
| 356 |
+
"loss": 0.9862,
|
| 357 |
+
"step": 86
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 5.176470588235294,
|
| 361 |
+
"grad_norm": 0.6497894525527954,
|
| 362 |
+
"learning_rate": 3.450980392156863e-06,
|
| 363 |
+
"loss": 0.9873,
|
| 364 |
+
"step": 88
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 5.294117647058823,
|
| 368 |
+
"grad_norm": 0.5283099412918091,
|
| 369 |
+
"learning_rate": 3.529411764705883e-06,
|
| 370 |
+
"loss": 0.9488,
|
| 371 |
+
"step": 90
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 5.411764705882353,
|
| 375 |
+
"grad_norm": 0.4905494451522827,
|
| 376 |
+
"learning_rate": 3.6078431372549024e-06,
|
| 377 |
+
"loss": 0.9606,
|
| 378 |
+
"step": 92
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.529411764705882,
|
| 382 |
+
"grad_norm": 0.5407238602638245,
|
| 383 |
+
"learning_rate": 3.6862745098039223e-06,
|
| 384 |
+
"loss": 0.888,
|
| 385 |
+
"step": 94
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 5.647058823529412,
|
| 389 |
+
"grad_norm": 0.5735270977020264,
|
| 390 |
+
"learning_rate": 3.7647058823529414e-06,
|
| 391 |
+
"loss": 0.916,
|
| 392 |
+
"step": 96
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 5.764705882352941,
|
| 396 |
+
"grad_norm": 0.5201605558395386,
|
| 397 |
+
"learning_rate": 3.843137254901962e-06,
|
| 398 |
+
"loss": 0.982,
|
| 399 |
+
"step": 98
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 5.882352941176471,
|
| 403 |
+
"grad_norm": 0.4470115303993225,
|
| 404 |
+
"learning_rate": 3.92156862745098e-06,
|
| 405 |
+
"loss": 0.9389,
|
| 406 |
+
"step": 100
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 6.0,
|
| 410 |
+
"grad_norm": 0.3844221532344818,
|
| 411 |
+
"learning_rate": 4.000000000000001e-06,
|
| 412 |
+
"loss": 1.0008,
|
| 413 |
+
"step": 102
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 6.0,
|
| 417 |
+
"eval_loss": 0.937701940536499,
|
| 418 |
+
"eval_runtime": 37.0036,
|
| 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": 0.4652542173862457,
|
| 426 |
+
"learning_rate": 4.07843137254902e-06,
|
| 427 |
+
"loss": 0.915,
|
| 428 |
+
"step": 104
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.235294117647059,
|
| 432 |
+
"grad_norm": 0.4144337773323059,
|
| 433 |
+
"learning_rate": 4.15686274509804e-06,
|
| 434 |
+
"loss": 0.8833,
|
| 435 |
+
"step": 106
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 6.352941176470588,
|
| 439 |
+
"grad_norm": 0.4218851327896118,
|
| 440 |
+
"learning_rate": 4.235294117647059e-06,
|
| 441 |
+
"loss": 0.8672,
|
| 442 |
+
"step": 108
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 6.470588235294118,
|
| 446 |
+
"grad_norm": 0.4070567786693573,
|
| 447 |
+
"learning_rate": 4.313725490196079e-06,
|
| 448 |
+
"loss": 0.9036,
|
| 449 |
+
"step": 110
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 6.588235294117647,
|
| 453 |
+
"grad_norm": 0.36765947937965393,
|
| 454 |
+
"learning_rate": 4.392156862745098e-06,
|
| 455 |
+
"loss": 0.9384,
|
| 456 |
+
"step": 112
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 6.705882352941177,
|
| 460 |
+
"grad_norm": 0.3613109886646271,
|
| 461 |
+
"learning_rate": 4.4705882352941184e-06,
|
| 462 |
+
"loss": 0.904,
|
| 463 |
+
"step": 114
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 6.823529411764706,
|
| 467 |
+
"grad_norm": 0.36222243309020996,
|
| 468 |
+
"learning_rate": 4.549019607843138e-06,
|
| 469 |
+
"loss": 0.8641,
|
| 470 |
+
"step": 116
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 6.9411764705882355,
|
| 474 |
+
"grad_norm": 0.42111915349960327,
|
| 475 |
+
"learning_rate": 4.627450980392157e-06,
|
| 476 |
+
"loss": 0.8935,
|
| 477 |
+
"step": 118
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 7.0,
|
| 481 |
+
"eval_loss": 0.8938563466072083,
|
| 482 |
+
"eval_runtime": 37.0541,
|
| 483 |
+
"eval_samples_per_second": 0.972,
|
| 484 |
+
"eval_steps_per_second": 0.972,
|
| 485 |
+
"step": 119
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 7.0588235294117645,
|
| 489 |
+
"grad_norm": 0.35050034523010254,
|
| 490 |
+
"learning_rate": 4.705882352941177e-06,
|
| 491 |
+
"loss": 0.9626,
|
| 492 |
+
"step": 120
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.176470588235294,
|
| 496 |
+
"grad_norm": 0.37680429220199585,
|
| 497 |
+
"learning_rate": 4.784313725490196e-06,
|
| 498 |
+
"loss": 0.8818,
|
| 499 |
+
"step": 122
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 7.294117647058823,
|
| 503 |
+
"grad_norm": 0.3310965895652771,
|
| 504 |
+
"learning_rate": 4.862745098039216e-06,
|
| 505 |
+
"loss": 0.8631,
|
| 506 |
+
"step": 124
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.411764705882353,
|
| 510 |
+
"grad_norm": 0.3843957781791687,
|
| 511 |
+
"learning_rate": 4.941176470588236e-06,
|
| 512 |
+
"loss": 0.8506,
|
| 513 |
+
"step": 126
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.529411764705882,
|
| 517 |
+
"grad_norm": 0.3577839732170105,
|
| 518 |
+
"learning_rate": 5.019607843137255e-06,
|
| 519 |
+
"loss": 0.841,
|
| 520 |
+
"step": 128
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.647058823529412,
|
| 524 |
+
"grad_norm": 0.3117690682411194,
|
| 525 |
+
"learning_rate": 5.098039215686274e-06,
|
| 526 |
+
"loss": 0.8576,
|
| 527 |
+
"step": 130
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 7.764705882352941,
|
| 531 |
+
"grad_norm": 0.37441667914390564,
|
| 532 |
+
"learning_rate": 5.176470588235295e-06,
|
| 533 |
+
"loss": 0.8127,
|
| 534 |
+
"step": 132
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 7.882352941176471,
|
| 538 |
+
"grad_norm": 0.3178256154060364,
|
| 539 |
+
"learning_rate": 5.254901960784314e-06,
|
| 540 |
+
"loss": 0.8554,
|
| 541 |
+
"step": 134
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 8.0,
|
| 545 |
+
"grad_norm": 0.3216207027435303,
|
| 546 |
+
"learning_rate": 5.333333333333334e-06,
|
| 547 |
+
"loss": 0.8862,
|
| 548 |
+
"step": 136
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 8.0,
|
| 552 |
+
"eval_loss": 0.8579282760620117,
|
| 553 |
+
"eval_runtime": 37.0044,
|
| 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": 0.341949462890625,
|
| 561 |
+
"learning_rate": 5.411764705882353e-06,
|
| 562 |
+
"loss": 0.8519,
|
| 563 |
+
"step": 138
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 8.235294117647058,
|
| 567 |
+
"grad_norm": 0.3150944113731384,
|
| 568 |
+
"learning_rate": 5.4901960784313735e-06,
|
| 569 |
+
"loss": 0.8338,
|
| 570 |
+
"step": 140
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 8.352941176470589,
|
| 574 |
+
"grad_norm": 0.3646303713321686,
|
| 575 |
+
"learning_rate": 5.568627450980393e-06,
|
| 576 |
+
"loss": 0.8381,
|
| 577 |
+
"step": 142
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 8.470588235294118,
|
| 581 |
+
"grad_norm": 0.334943950176239,
|
| 582 |
+
"learning_rate": 5.6470588235294125e-06,
|
| 583 |
+
"loss": 0.879,
|
| 584 |
+
"step": 144
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 8.588235294117647,
|
| 588 |
+
"grad_norm": 0.3029642403125763,
|
| 589 |
+
"learning_rate": 5.725490196078431e-06,
|
| 590 |
+
"loss": 0.7807,
|
| 591 |
+
"step": 146
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 8.705882352941176,
|
| 595 |
+
"grad_norm": 0.30906420946121216,
|
| 596 |
+
"learning_rate": 5.803921568627452e-06,
|
| 597 |
+
"loss": 0.7946,
|
| 598 |
+
"step": 148
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 8.823529411764707,
|
| 602 |
+
"grad_norm": 0.34527918696403503,
|
| 603 |
+
"learning_rate": 5.882352941176471e-06,
|
| 604 |
+
"loss": 0.8397,
|
| 605 |
+
"step": 150
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 8.941176470588236,
|
| 609 |
+
"grad_norm": 0.3913155794143677,
|
| 610 |
+
"learning_rate": 5.96078431372549e-06,
|
| 611 |
+
"loss": 0.8266,
|
| 612 |
+
"step": 152
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 9.0,
|
| 616 |
+
"eval_loss": 0.8293797969818115,
|
| 617 |
+
"eval_runtime": 37.0023,
|
| 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": 0.3411412835121155,
|
| 625 |
+
"learning_rate": 6.03921568627451e-06,
|
| 626 |
+
"loss": 0.8249,
|
| 627 |
+
"step": 154
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 9.176470588235293,
|
| 631 |
+
"grad_norm": 0.28964897990226746,
|
| 632 |
+
"learning_rate": 6.11764705882353e-06,
|
| 633 |
+
"loss": 0.7887,
|
| 634 |
+
"step": 156
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 9.294117647058824,
|
| 638 |
+
"grad_norm": 0.36540400981903076,
|
| 639 |
+
"learning_rate": 6.19607843137255e-06,
|
| 640 |
+
"loss": 0.7434,
|
| 641 |
+
"step": 158
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 9.411764705882353,
|
| 645 |
+
"grad_norm": 0.3152979016304016,
|
| 646 |
+
"learning_rate": 6.274509803921569e-06,
|
| 647 |
+
"loss": 0.7864,
|
| 648 |
+
"step": 160
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 9.529411764705882,
|
| 652 |
+
"grad_norm": 0.28152012825012207,
|
| 653 |
+
"learning_rate": 6.352941176470589e-06,
|
| 654 |
+
"loss": 0.777,
|
| 655 |
+
"step": 162
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 9.647058823529411,
|
| 659 |
+
"grad_norm": 0.37814223766326904,
|
| 660 |
+
"learning_rate": 6.431372549019609e-06,
|
| 661 |
+
"loss": 0.8319,
|
| 662 |
+
"step": 164
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 9.764705882352942,
|
| 666 |
+
"grad_norm": 0.3085719645023346,
|
| 667 |
+
"learning_rate": 6.5098039215686285e-06,
|
| 668 |
+
"loss": 0.7986,
|
| 669 |
+
"step": 166
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 9.882352941176471,
|
| 673 |
+
"grad_norm": 0.29185226559638977,
|
| 674 |
+
"learning_rate": 6.588235294117647e-06,
|
| 675 |
+
"loss": 0.8292,
|
| 676 |
+
"step": 168
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 10.0,
|
| 680 |
+
"grad_norm": 0.30540931224823,
|
| 681 |
+
"learning_rate": 6.666666666666667e-06,
|
| 682 |
+
"loss": 0.7797,
|
| 683 |
+
"step": 170
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 10.0,
|
| 687 |
+
"eval_loss": 0.807489275932312,
|
| 688 |
+
"eval_runtime": 37.0049,
|
| 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": 0.3091186285018921,
|
| 696 |
+
"learning_rate": 6.745098039215687e-06,
|
| 697 |
+
"loss": 0.7765,
|
| 698 |
+
"step": 172
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 10.235294117647058,
|
| 702 |
+
"grad_norm": 0.30727335810661316,
|
| 703 |
+
"learning_rate": 6.8235294117647065e-06,
|
| 704 |
+
"loss": 0.7505,
|
| 705 |
+
"step": 174
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 10.352941176470589,
|
| 709 |
+
"grad_norm": 0.2999902665615082,
|
| 710 |
+
"learning_rate": 6.901960784313726e-06,
|
| 711 |
+
"loss": 0.7233,
|
| 712 |
+
"step": 176
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 10.470588235294118,
|
| 716 |
+
"grad_norm": 0.2925303876399994,
|
| 717 |
+
"learning_rate": 6.9803921568627454e-06,
|
| 718 |
+
"loss": 0.7613,
|
| 719 |
+
"step": 178
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 10.588235294117647,
|
| 723 |
+
"grad_norm": 0.28137609362602234,
|
| 724 |
+
"learning_rate": 7.058823529411766e-06,
|
| 725 |
+
"loss": 0.7336,
|
| 726 |
+
"step": 180
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 10.705882352941176,
|
| 730 |
+
"grad_norm": 0.3438097834587097,
|
| 731 |
+
"learning_rate": 7.137254901960785e-06,
|
| 732 |
+
"loss": 0.822,
|
| 733 |
+
"step": 182
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 10.823529411764707,
|
| 737 |
+
"grad_norm": 0.3309740722179413,
|
| 738 |
+
"learning_rate": 7.215686274509805e-06,
|
| 739 |
+
"loss": 0.8488,
|
| 740 |
+
"step": 184
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 10.941176470588236,
|
| 744 |
+
"grad_norm": 0.33822691440582275,
|
| 745 |
+
"learning_rate": 7.294117647058823e-06,
|
| 746 |
+
"loss": 0.8158,
|
| 747 |
+
"step": 186
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 11.0,
|
| 751 |
+
"eval_loss": 0.790329098701477,
|
| 752 |
+
"eval_runtime": 37.0086,
|
| 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": 0.3196800649166107,
|
| 760 |
+
"learning_rate": 7.372549019607845e-06,
|
| 761 |
+
"loss": 0.7996,
|
| 762 |
+
"step": 188
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 11.176470588235293,
|
| 766 |
+
"grad_norm": 0.2967463731765747,
|
| 767 |
+
"learning_rate": 7.450980392156863e-06,
|
| 768 |
+
"loss": 0.7102,
|
| 769 |
+
"step": 190
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 11.294117647058824,
|
| 773 |
+
"grad_norm": 0.32051464915275574,
|
| 774 |
+
"learning_rate": 7.529411764705883e-06,
|
| 775 |
+
"loss": 0.8008,
|
| 776 |
+
"step": 192
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 11.411764705882353,
|
| 780 |
+
"grad_norm": 0.399475634098053,
|
| 781 |
+
"learning_rate": 7.607843137254902e-06,
|
| 782 |
+
"loss": 0.8271,
|
| 783 |
+
"step": 194
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 11.529411764705882,
|
| 787 |
+
"grad_norm": 0.3100043535232544,
|
| 788 |
+
"learning_rate": 7.686274509803923e-06,
|
| 789 |
+
"loss": 0.7954,
|
| 790 |
+
"step": 196
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 11.647058823529411,
|
| 794 |
+
"grad_norm": 0.31522953510284424,
|
| 795 |
+
"learning_rate": 7.764705882352941e-06,
|
| 796 |
+
"loss": 0.712,
|
| 797 |
+
"step": 198
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 11.764705882352942,
|
| 801 |
+
"grad_norm": 0.3450244665145874,
|
| 802 |
+
"learning_rate": 7.84313725490196e-06,
|
| 803 |
+
"loss": 0.7329,
|
| 804 |
+
"step": 200
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 11.882352941176471,
|
| 808 |
+
"grad_norm": 0.3267311453819275,
|
| 809 |
+
"learning_rate": 7.92156862745098e-06,
|
| 810 |
+
"loss": 0.7275,
|
| 811 |
+
"step": 202
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 12.0,
|
| 815 |
+
"grad_norm": 0.42111936211586,
|
| 816 |
+
"learning_rate": 8.000000000000001e-06,
|
| 817 |
+
"loss": 0.6845,
|
| 818 |
+
"step": 204
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 12.0,
|
| 822 |
+
"eval_loss": 0.7741987705230713,
|
| 823 |
+
"eval_runtime": 37.0028,
|
| 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": 0.33904674649238586,
|
| 831 |
+
"learning_rate": 8.07843137254902e-06,
|
| 832 |
+
"loss": 0.7635,
|
| 833 |
+
"step": 206
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 12.235294117647058,
|
| 837 |
+
"grad_norm": 0.33812686800956726,
|
| 838 |
+
"learning_rate": 8.15686274509804e-06,
|
| 839 |
+
"loss": 0.8239,
|
| 840 |
+
"step": 208
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 12.352941176470589,
|
| 844 |
+
"grad_norm": 0.34535664319992065,
|
| 845 |
+
"learning_rate": 8.23529411764706e-06,
|
| 846 |
+
"loss": 0.7084,
|
| 847 |
+
"step": 210
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 12.470588235294118,
|
| 851 |
+
"grad_norm": 0.330837219953537,
|
| 852 |
+
"learning_rate": 8.31372549019608e-06,
|
| 853 |
+
"loss": 0.6769,
|
| 854 |
+
"step": 212
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 12.588235294117647,
|
| 858 |
+
"grad_norm": 0.31368908286094666,
|
| 859 |
+
"learning_rate": 8.392156862745099e-06,
|
| 860 |
+
"loss": 0.7651,
|
| 861 |
+
"step": 214
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 12.705882352941176,
|
| 865 |
+
"grad_norm": 0.35061123967170715,
|
| 866 |
+
"learning_rate": 8.470588235294118e-06,
|
| 867 |
+
"loss": 0.7452,
|
| 868 |
+
"step": 216
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 12.823529411764707,
|
| 872 |
+
"grad_norm": 0.32111451029777527,
|
| 873 |
+
"learning_rate": 8.549019607843138e-06,
|
| 874 |
+
"loss": 0.718,
|
| 875 |
+
"step": 218
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 12.941176470588236,
|
| 879 |
+
"grad_norm": 0.3150181174278259,
|
| 880 |
+
"learning_rate": 8.627450980392157e-06,
|
| 881 |
+
"loss": 0.6819,
|
| 882 |
+
"step": 220
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 13.0,
|
| 886 |
+
"eval_loss": 0.7598351240158081,
|
| 887 |
+
"eval_runtime": 37.0068,
|
| 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": 0.3078894019126892,
|
| 895 |
+
"learning_rate": 8.705882352941177e-06,
|
| 896 |
+
"loss": 0.7151,
|
| 897 |
+
"step": 222
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 13.176470588235293,
|
| 901 |
+
"grad_norm": 0.3273066282272339,
|
| 902 |
+
"learning_rate": 8.784313725490196e-06,
|
| 903 |
+
"loss": 0.6766,
|
| 904 |
+
"step": 224
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 13.294117647058824,
|
| 908 |
+
"grad_norm": 0.30240604281425476,
|
| 909 |
+
"learning_rate": 8.862745098039216e-06,
|
| 910 |
+
"loss": 0.665,
|
| 911 |
+
"step": 226
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 13.411764705882353,
|
| 915 |
+
"grad_norm": 0.3365361988544464,
|
| 916 |
+
"learning_rate": 8.941176470588237e-06,
|
| 917 |
+
"loss": 0.73,
|
| 918 |
+
"step": 228
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 13.529411764705882,
|
| 922 |
+
"grad_norm": 0.3776240646839142,
|
| 923 |
+
"learning_rate": 9.019607843137256e-06,
|
| 924 |
+
"loss": 0.7319,
|
| 925 |
+
"step": 230
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 13.647058823529411,
|
| 929 |
+
"grad_norm": 0.32826074957847595,
|
| 930 |
+
"learning_rate": 9.098039215686276e-06,
|
| 931 |
+
"loss": 0.7021,
|
| 932 |
+
"step": 232
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 13.764705882352942,
|
| 936 |
+
"grad_norm": 0.3192596435546875,
|
| 937 |
+
"learning_rate": 9.176470588235294e-06,
|
| 938 |
+
"loss": 0.7333,
|
| 939 |
+
"step": 234
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 13.882352941176471,
|
| 943 |
+
"grad_norm": 0.36111271381378174,
|
| 944 |
+
"learning_rate": 9.254901960784315e-06,
|
| 945 |
+
"loss": 0.7633,
|
| 946 |
+
"step": 236
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 14.0,
|
| 950 |
+
"grad_norm": 0.3694405257701874,
|
| 951 |
+
"learning_rate": 9.333333333333334e-06,
|
| 952 |
+
"loss": 0.7241,
|
| 953 |
+
"step": 238
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 14.0,
|
| 957 |
+
"eval_loss": 0.7471711039543152,
|
| 958 |
+
"eval_runtime": 37.0047,
|
| 959 |
+
"eval_samples_per_second": 0.973,
|
| 960 |
+
"eval_steps_per_second": 0.973,
|
| 961 |
+
"step": 238
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 14.117647058823529,
|
| 965 |
+
"grad_norm": 0.3688935339450836,
|
| 966 |
+
"learning_rate": 9.411764705882354e-06,
|
| 967 |
+
"loss": 0.742,
|
| 968 |
+
"step": 240
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 14.235294117647058,
|
| 972 |
+
"grad_norm": 0.41638877987861633,
|
| 973 |
+
"learning_rate": 9.490196078431373e-06,
|
| 974 |
+
"loss": 0.7049,
|
| 975 |
+
"step": 242
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 14.352941176470589,
|
| 979 |
+
"grad_norm": 0.3489099442958832,
|
| 980 |
+
"learning_rate": 9.568627450980393e-06,
|
| 981 |
+
"loss": 0.6517,
|
| 982 |
+
"step": 244
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 14.470588235294118,
|
| 986 |
+
"grad_norm": 0.3671415448188782,
|
| 987 |
+
"learning_rate": 9.647058823529412e-06,
|
| 988 |
+
"loss": 0.7233,
|
| 989 |
+
"step": 246
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 14.588235294117647,
|
| 993 |
+
"grad_norm": 0.37552204728126526,
|
| 994 |
+
"learning_rate": 9.725490196078432e-06,
|
| 995 |
+
"loss": 0.701,
|
| 996 |
+
"step": 248
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 14.705882352941176,
|
| 1000 |
+
"grad_norm": 0.43261757493019104,
|
| 1001 |
+
"learning_rate": 9.803921568627451e-06,
|
| 1002 |
+
"loss": 0.6979,
|
| 1003 |
+
"step": 250
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 14.823529411764707,
|
| 1007 |
+
"grad_norm": 0.37599849700927734,
|
| 1008 |
+
"learning_rate": 9.882352941176472e-06,
|
| 1009 |
+
"loss": 0.6462,
|
| 1010 |
+
"step": 252
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 14.941176470588236,
|
| 1014 |
+
"grad_norm": 0.3583243191242218,
|
| 1015 |
+
"learning_rate": 9.960784313725492e-06,
|
| 1016 |
+
"loss": 0.695,
|
| 1017 |
+
"step": 254
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 15.0,
|
| 1021 |
+
"eval_loss": 0.7365108132362366,
|
| 1022 |
+
"eval_runtime": 37.003,
|
| 1023 |
+
"eval_samples_per_second": 0.973,
|
| 1024 |
+
"eval_steps_per_second": 0.973,
|
| 1025 |
+
"step": 255
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 15.058823529411764,
|
| 1029 |
+
"grad_norm": 0.4206070005893707,
|
| 1030 |
+
"learning_rate": 9.999995315380667e-06,
|
| 1031 |
+
"loss": 0.6862,
|
| 1032 |
+
"step": 256
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 15.176470588235293,
|
| 1036 |
+
"grad_norm": 0.47259286046028137,
|
| 1037 |
+
"learning_rate": 9.99995783847866e-06,
|
| 1038 |
+
"loss": 0.6933,
|
| 1039 |
+
"step": 258
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 15.294117647058824,
|
| 1043 |
+
"grad_norm": 0.3962372839450836,
|
| 1044 |
+
"learning_rate": 9.999882884955554e-06,
|
| 1045 |
+
"loss": 0.6643,
|
| 1046 |
+
"step": 260
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 15.411764705882353,
|
| 1050 |
+
"grad_norm": 0.37228551506996155,
|
| 1051 |
+
"learning_rate": 9.99977045537315e-06,
|
| 1052 |
+
"loss": 0.6676,
|
| 1053 |
+
"step": 262
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 15.529411764705882,
|
| 1057 |
+
"grad_norm": 0.4293304979801178,
|
| 1058 |
+
"learning_rate": 9.999620550574155e-06,
|
| 1059 |
+
"loss": 0.6497,
|
| 1060 |
+
"step": 264
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 15.647058823529411,
|
| 1064 |
+
"grad_norm": 0.4316819906234741,
|
| 1065 |
+
"learning_rate": 9.999433171682158e-06,
|
| 1066 |
+
"loss": 0.6687,
|
| 1067 |
+
"step": 266
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 15.764705882352942,
|
| 1071 |
+
"grad_norm": 0.4474015533924103,
|
| 1072 |
+
"learning_rate": 9.999208320101643e-06,
|
| 1073 |
+
"loss": 0.6935,
|
| 1074 |
+
"step": 268
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 15.882352941176471,
|
| 1078 |
+
"grad_norm": 0.4514879584312439,
|
| 1079 |
+
"learning_rate": 9.998945997517957e-06,
|
| 1080 |
+
"loss": 0.6542,
|
| 1081 |
+
"step": 270
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 16.0,
|
| 1085 |
+
"grad_norm": 0.5015867948532104,
|
| 1086 |
+
"learning_rate": 9.99864620589731e-06,
|
| 1087 |
+
"loss": 0.6982,
|
| 1088 |
+
"step": 272
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 16.0,
|
| 1092 |
+
"eval_loss": 0.7271888256072998,
|
| 1093 |
+
"eval_runtime": 37.0158,
|
| 1094 |
+
"eval_samples_per_second": 0.973,
|
| 1095 |
+
"eval_steps_per_second": 0.973,
|
| 1096 |
+
"step": 272
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 16.11764705882353,
|
| 1100 |
+
"grad_norm": 0.5128858685493469,
|
| 1101 |
+
"learning_rate": 9.998308947486753e-06,
|
| 1102 |
+
"loss": 0.6414,
|
| 1103 |
+
"step": 274
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 16.235294117647058,
|
| 1107 |
+
"grad_norm": 0.40119990706443787,
|
| 1108 |
+
"learning_rate": 9.997934224814173e-06,
|
| 1109 |
+
"loss": 0.628,
|
| 1110 |
+
"step": 276
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 16.352941176470587,
|
| 1114 |
+
"grad_norm": 0.4133683145046234,
|
| 1115 |
+
"learning_rate": 9.997522040688258e-06,
|
| 1116 |
+
"loss": 0.7072,
|
| 1117 |
+
"step": 278
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 16.470588235294116,
|
| 1121 |
+
"grad_norm": 0.41614237427711487,
|
| 1122 |
+
"learning_rate": 9.997072398198492e-06,
|
| 1123 |
+
"loss": 0.6526,
|
| 1124 |
+
"step": 280
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 16.58823529411765,
|
| 1128 |
+
"grad_norm": 0.4795832931995392,
|
| 1129 |
+
"learning_rate": 9.996585300715117e-06,
|
| 1130 |
+
"loss": 0.6583,
|
| 1131 |
+
"step": 282
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 16.705882352941178,
|
| 1135 |
+
"grad_norm": 0.4348108470439911,
|
| 1136 |
+
"learning_rate": 9.996060751889114e-06,
|
| 1137 |
+
"loss": 0.6728,
|
| 1138 |
+
"step": 284
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 16.823529411764707,
|
| 1142 |
+
"grad_norm": 0.4476607143878937,
|
| 1143 |
+
"learning_rate": 9.995498755652186e-06,
|
| 1144 |
+
"loss": 0.626,
|
| 1145 |
+
"step": 286
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 16.941176470588236,
|
| 1149 |
+
"grad_norm": 0.4528913199901581,
|
| 1150 |
+
"learning_rate": 9.994899316216709e-06,
|
| 1151 |
+
"loss": 0.622,
|
| 1152 |
+
"step": 288
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 17.0,
|
| 1156 |
+
"eval_loss": 0.7215262055397034,
|
| 1157 |
+
"eval_runtime": 36.9993,
|
| 1158 |
+
"eval_samples_per_second": 0.973,
|
| 1159 |
+
"eval_steps_per_second": 0.973,
|
| 1160 |
+
"step": 289
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 17.058823529411764,
|
| 1164 |
+
"grad_norm": 0.5009466409683228,
|
| 1165 |
+
"learning_rate": 9.994262438075713e-06,
|
| 1166 |
+
"loss": 0.7081,
|
| 1167 |
+
"step": 290
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 17.176470588235293,
|
| 1171 |
+
"grad_norm": 0.475374698638916,
|
| 1172 |
+
"learning_rate": 9.993588126002848e-06,
|
| 1173 |
+
"loss": 0.6229,
|
| 1174 |
+
"step": 292
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 17.294117647058822,
|
| 1178 |
+
"grad_norm": 0.4642236828804016,
|
| 1179 |
+
"learning_rate": 9.992876385052346e-06,
|
| 1180 |
+
"loss": 0.5844,
|
| 1181 |
+
"step": 294
|
| 1182 |
+
},
|
| 1183 |
+
{
|
| 1184 |
+
"epoch": 17.41176470588235,
|
| 1185 |
+
"grad_norm": 0.4536580741405487,
|
| 1186 |
+
"learning_rate": 9.992127220558976e-06,
|
| 1187 |
+
"loss": 0.6135,
|
| 1188 |
+
"step": 296
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 17.529411764705884,
|
| 1192 |
+
"grad_norm": 0.45987674593925476,
|
| 1193 |
+
"learning_rate": 9.991340638138022e-06,
|
| 1194 |
+
"loss": 0.6511,
|
| 1195 |
+
"step": 298
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 17.647058823529413,
|
| 1199 |
+
"grad_norm": 0.4978199899196625,
|
| 1200 |
+
"learning_rate": 9.990516643685222e-06,
|
| 1201 |
+
"loss": 0.6424,
|
| 1202 |
+
"step": 300
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 17.764705882352942,
|
| 1206 |
+
"grad_norm": 0.49036839604377747,
|
| 1207 |
+
"learning_rate": 9.98965524337673e-06,
|
| 1208 |
+
"loss": 0.6773,
|
| 1209 |
+
"step": 302
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 17.88235294117647,
|
| 1213 |
+
"grad_norm": 0.5300158858299255,
|
| 1214 |
+
"learning_rate": 9.988756443669081e-06,
|
| 1215 |
+
"loss": 0.6364,
|
| 1216 |
+
"step": 304
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 18.0,
|
| 1220 |
+
"grad_norm": 0.4950679540634155,
|
| 1221 |
+
"learning_rate": 9.987820251299121e-06,
|
| 1222 |
+
"loss": 0.5905,
|
| 1223 |
+
"step": 306
|
| 1224 |
+
},
|
| 1225 |
+
{
|
| 1226 |
+
"epoch": 18.0,
|
| 1227 |
+
"eval_loss": 0.7155699133872986,
|
| 1228 |
+
"eval_runtime": 37.0042,
|
| 1229 |
+
"eval_samples_per_second": 0.973,
|
| 1230 |
+
"eval_steps_per_second": 0.973,
|
| 1231 |
+
"step": 306
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 18.11764705882353,
|
| 1235 |
+
"grad_norm": 0.5069499611854553,
|
| 1236 |
+
"learning_rate": 9.98684667328398e-06,
|
| 1237 |
+
"loss": 0.6123,
|
| 1238 |
+
"step": 308
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 18.235294117647058,
|
| 1242 |
+
"grad_norm": 0.46550482511520386,
|
| 1243 |
+
"learning_rate": 9.985835716921e-06,
|
| 1244 |
+
"loss": 0.6102,
|
| 1245 |
+
"step": 310
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 18.352941176470587,
|
| 1249 |
+
"grad_norm": 0.7860422730445862,
|
| 1250 |
+
"learning_rate": 9.984787389787689e-06,
|
| 1251 |
+
"loss": 0.6273,
|
| 1252 |
+
"step": 312
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 18.470588235294116,
|
| 1256 |
+
"grad_norm": 0.49825742840766907,
|
| 1257 |
+
"learning_rate": 9.983701699741668e-06,
|
| 1258 |
+
"loss": 0.6297,
|
| 1259 |
+
"step": 314
|
| 1260 |
+
},
|
| 1261 |
+
{
|
| 1262 |
+
"epoch": 18.58823529411765,
|
| 1263 |
+
"grad_norm": 0.5290573239326477,
|
| 1264 |
+
"learning_rate": 9.982578654920601e-06,
|
| 1265 |
+
"loss": 0.6238,
|
| 1266 |
+
"step": 316
|
| 1267 |
+
},
|
| 1268 |
+
{
|
| 1269 |
+
"epoch": 18.705882352941178,
|
| 1270 |
+
"grad_norm": 0.5185476541519165,
|
| 1271 |
+
"learning_rate": 9.981418263742148e-06,
|
| 1272 |
+
"loss": 0.5702,
|
| 1273 |
+
"step": 318
|
| 1274 |
+
},
|
| 1275 |
+
{
|
| 1276 |
+
"epoch": 18.823529411764707,
|
| 1277 |
+
"grad_norm": 0.5360794067382812,
|
| 1278 |
+
"learning_rate": 9.980220534903889e-06,
|
| 1279 |
+
"loss": 0.6082,
|
| 1280 |
+
"step": 320
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 18.941176470588236,
|
| 1284 |
+
"grad_norm": 0.5598728656768799,
|
| 1285 |
+
"learning_rate": 9.978985477383264e-06,
|
| 1286 |
+
"loss": 0.6121,
|
| 1287 |
+
"step": 322
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 19.0,
|
| 1291 |
+
"eval_loss": 0.7140286564826965,
|
| 1292 |
+
"eval_runtime": 36.9943,
|
| 1293 |
+
"eval_samples_per_second": 0.973,
|
| 1294 |
+
"eval_steps_per_second": 0.973,
|
| 1295 |
+
"step": 323
|
| 1296 |
+
},
|
| 1297 |
+
{
|
| 1298 |
+
"epoch": 19.058823529411764,
|
| 1299 |
+
"grad_norm": 0.5005812048912048,
|
| 1300 |
+
"learning_rate": 9.97771310043751e-06,
|
| 1301 |
+
"loss": 0.5964,
|
| 1302 |
+
"step": 324
|
| 1303 |
+
},
|
| 1304 |
+
{
|
| 1305 |
+
"epoch": 19.176470588235293,
|
| 1306 |
+
"grad_norm": 0.5243679881095886,
|
| 1307 |
+
"learning_rate": 9.97640341360358e-06,
|
| 1308 |
+
"loss": 0.643,
|
| 1309 |
+
"step": 326
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 19.294117647058822,
|
| 1313 |
+
"grad_norm": 0.5746588706970215,
|
| 1314 |
+
"learning_rate": 9.975056426698094e-06,
|
| 1315 |
+
"loss": 0.587,
|
| 1316 |
+
"step": 328
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 19.41176470588235,
|
| 1320 |
+
"grad_norm": 0.5213414430618286,
|
| 1321 |
+
"learning_rate": 9.973672149817232e-06,
|
| 1322 |
+
"loss": 0.5835,
|
| 1323 |
+
"step": 330
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 19.529411764705884,
|
| 1327 |
+
"grad_norm": 0.6261019706726074,
|
| 1328 |
+
"learning_rate": 9.972250593336689e-06,
|
| 1329 |
+
"loss": 0.5884,
|
| 1330 |
+
"step": 332
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 19.647058823529413,
|
| 1334 |
+
"grad_norm": 0.5905741453170776,
|
| 1335 |
+
"learning_rate": 9.970791767911581e-06,
|
| 1336 |
+
"loss": 0.588,
|
| 1337 |
+
"step": 334
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 19.764705882352942,
|
| 1341 |
+
"grad_norm": 0.5681843757629395,
|
| 1342 |
+
"learning_rate": 9.96929568447637e-06,
|
| 1343 |
+
"loss": 0.554,
|
| 1344 |
+
"step": 336
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 19.88235294117647,
|
| 1348 |
+
"grad_norm": 0.6130782961845398,
|
| 1349 |
+
"learning_rate": 9.967762354244778e-06,
|
| 1350 |
+
"loss": 0.6003,
|
| 1351 |
+
"step": 338
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 20.0,
|
| 1355 |
+
"grad_norm": 0.6039074659347534,
|
| 1356 |
+
"learning_rate": 9.966191788709716e-06,
|
| 1357 |
+
"loss": 0.567,
|
| 1358 |
+
"step": 340
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 20.0,
|
| 1362 |
+
"eval_loss": 0.7166150808334351,
|
| 1363 |
+
"eval_runtime": 36.9981,
|
| 1364 |
+
"eval_samples_per_second": 0.973,
|
| 1365 |
+
"eval_steps_per_second": 0.973,
|
| 1366 |
+
"step": 340
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 20.11764705882353,
|
| 1370 |
+
"grad_norm": 0.661340594291687,
|
| 1371 |
+
"learning_rate": 9.964583999643174e-06,
|
| 1372 |
+
"loss": 0.5771,
|
| 1373 |
+
"step": 342
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 20.235294117647058,
|
| 1377 |
+
"grad_norm": 0.6773268580436707,
|
| 1378 |
+
"learning_rate": 9.962938999096159e-06,
|
| 1379 |
+
"loss": 0.5611,
|
| 1380 |
+
"step": 344
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 20.352941176470587,
|
| 1384 |
+
"grad_norm": 0.6907221674919128,
|
| 1385 |
+
"learning_rate": 9.961256799398584e-06,
|
| 1386 |
+
"loss": 0.5961,
|
| 1387 |
+
"step": 346
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 20.470588235294116,
|
| 1391 |
+
"grad_norm": 0.6799242496490479,
|
| 1392 |
+
"learning_rate": 9.95953741315919e-06,
|
| 1393 |
+
"loss": 0.605,
|
| 1394 |
+
"step": 348
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 20.58823529411765,
|
| 1398 |
+
"grad_norm": 0.6507272720336914,
|
| 1399 |
+
"learning_rate": 9.957780853265441e-06,
|
| 1400 |
+
"loss": 0.5564,
|
| 1401 |
+
"step": 350
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 20.705882352941178,
|
| 1405 |
+
"grad_norm": 0.7043225765228271,
|
| 1406 |
+
"learning_rate": 9.955987132883435e-06,
|
| 1407 |
+
"loss": 0.5069,
|
| 1408 |
+
"step": 352
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 20.823529411764707,
|
| 1412 |
+
"grad_norm": 0.7170704007148743,
|
| 1413 |
+
"learning_rate": 9.954156265457801e-06,
|
| 1414 |
+
"loss": 0.5671,
|
| 1415 |
+
"step": 354
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 20.941176470588236,
|
| 1419 |
+
"grad_norm": 0.6098182797431946,
|
| 1420 |
+
"learning_rate": 9.952288264711601e-06,
|
| 1421 |
+
"loss": 0.5471,
|
| 1422 |
+
"step": 356
|
| 1423 |
+
},
|
| 1424 |
+
{
|
| 1425 |
+
"epoch": 21.0,
|
| 1426 |
+
"eval_loss": 0.7171775698661804,
|
| 1427 |
+
"eval_runtime": 36.9995,
|
| 1428 |
+
"eval_samples_per_second": 0.973,
|
| 1429 |
+
"eval_steps_per_second": 0.973,
|
| 1430 |
+
"step": 357
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 21.058823529411764,
|
| 1434 |
+
"grad_norm": 0.7199786305427551,
|
| 1435 |
+
"learning_rate": 9.950383144646221e-06,
|
| 1436 |
+
"loss": 0.5856,
|
| 1437 |
+
"step": 358
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 21.176470588235293,
|
| 1441 |
+
"grad_norm": 0.748616635799408,
|
| 1442 |
+
"learning_rate": 9.948440919541277e-06,
|
| 1443 |
+
"loss": 0.5347,
|
| 1444 |
+
"step": 360
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 21.294117647058822,
|
| 1448 |
+
"grad_norm": 0.7035663723945618,
|
| 1449 |
+
"learning_rate": 9.946461603954499e-06,
|
| 1450 |
+
"loss": 0.5433,
|
| 1451 |
+
"step": 362
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 21.41176470588235,
|
| 1455 |
+
"grad_norm": 0.7293503880500793,
|
| 1456 |
+
"learning_rate": 9.944445212721619e-06,
|
| 1457 |
+
"loss": 0.5473,
|
| 1458 |
+
"step": 364
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 21.529411764705884,
|
| 1462 |
+
"grad_norm": 0.7407766580581665,
|
| 1463 |
+
"learning_rate": 9.942391760956277e-06,
|
| 1464 |
+
"loss": 0.527,
|
| 1465 |
+
"step": 366
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 21.647058823529413,
|
| 1469 |
+
"grad_norm": 0.7856358289718628,
|
| 1470 |
+
"learning_rate": 9.940301264049885e-06,
|
| 1471 |
+
"loss": 0.5372,
|
| 1472 |
+
"step": 368
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 21.764705882352942,
|
| 1476 |
+
"grad_norm": 0.7008533477783203,
|
| 1477 |
+
"learning_rate": 9.938173737671531e-06,
|
| 1478 |
+
"loss": 0.5375,
|
| 1479 |
+
"step": 370
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 21.88235294117647,
|
| 1483 |
+
"grad_norm": 0.7586143612861633,
|
| 1484 |
+
"learning_rate": 9.936009197767847e-06,
|
| 1485 |
+
"loss": 0.5467,
|
| 1486 |
+
"step": 372
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 22.0,
|
| 1490 |
+
"grad_norm": 0.8423267602920532,
|
| 1491 |
+
"learning_rate": 9.933807660562898e-06,
|
| 1492 |
+
"loss": 0.4761,
|
| 1493 |
+
"step": 374
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 22.0,
|
| 1497 |
+
"eval_loss": 0.7233606576919556,
|
| 1498 |
+
"eval_runtime": 37.0054,
|
| 1499 |
+
"eval_samples_per_second": 0.973,
|
| 1500 |
+
"eval_steps_per_second": 0.973,
|
| 1501 |
+
"step": 374
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 22.11764705882353,
|
| 1505 |
+
"grad_norm": 0.882168710231781,
|
| 1506 |
+
"learning_rate": 9.931569142558057e-06,
|
| 1507 |
+
"loss": 0.562,
|
| 1508 |
+
"step": 376
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 22.235294117647058,
|
| 1512 |
+
"grad_norm": 0.8616783618927002,
|
| 1513 |
+
"learning_rate": 9.929293660531889e-06,
|
| 1514 |
+
"loss": 0.5039,
|
| 1515 |
+
"step": 378
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 22.352941176470587,
|
| 1519 |
+
"grad_norm": 0.9095832705497742,
|
| 1520 |
+
"learning_rate": 9.926981231540007e-06,
|
| 1521 |
+
"loss": 0.4744,
|
| 1522 |
+
"step": 380
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 22.470588235294116,
|
| 1526 |
+
"grad_norm": 0.8473853468894958,
|
| 1527 |
+
"learning_rate": 9.924631872914967e-06,
|
| 1528 |
+
"loss": 0.4976,
|
| 1529 |
+
"step": 382
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 22.58823529411765,
|
| 1533 |
+
"grad_norm": 0.844835102558136,
|
| 1534 |
+
"learning_rate": 9.922245602266119e-06,
|
| 1535 |
+
"loss": 0.4721,
|
| 1536 |
+
"step": 384
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 22.705882352941178,
|
| 1540 |
+
"grad_norm": 0.820750892162323,
|
| 1541 |
+
"learning_rate": 9.919822437479488e-06,
|
| 1542 |
+
"loss": 0.4995,
|
| 1543 |
+
"step": 386
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 22.823529411764707,
|
| 1547 |
+
"grad_norm": 0.823648989200592,
|
| 1548 |
+
"learning_rate": 9.91736239671763e-06,
|
| 1549 |
+
"loss": 0.5061,
|
| 1550 |
+
"step": 388
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 22.941176470588236,
|
| 1554 |
+
"grad_norm": 0.8103899359703064,
|
| 1555 |
+
"learning_rate": 9.91486549841951e-06,
|
| 1556 |
+
"loss": 0.4967,
|
| 1557 |
+
"step": 390
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 23.0,
|
| 1561 |
+
"eval_loss": 0.7357540726661682,
|
| 1562 |
+
"eval_runtime": 37.0101,
|
| 1563 |
+
"eval_samples_per_second": 0.973,
|
| 1564 |
+
"eval_steps_per_second": 0.973,
|
| 1565 |
+
"step": 391
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 23.058823529411764,
|
| 1569 |
+
"grad_norm": 0.7696477174758911,
|
| 1570 |
+
"learning_rate": 9.912331761300341e-06,
|
| 1571 |
+
"loss": 0.5004,
|
| 1572 |
+
"step": 392
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 23.176470588235293,
|
| 1576 |
+
"grad_norm": 0.7803938984870911,
|
| 1577 |
+
"learning_rate": 9.909761204351469e-06,
|
| 1578 |
+
"loss": 0.4903,
|
| 1579 |
+
"step": 394
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 23.294117647058822,
|
| 1583 |
+
"grad_norm": 0.9065766930580139,
|
| 1584 |
+
"learning_rate": 9.90715384684021e-06,
|
| 1585 |
+
"loss": 0.505,
|
| 1586 |
+
"step": 396
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 23.41176470588235,
|
| 1590 |
+
"grad_norm": 1.1152830123901367,
|
| 1591 |
+
"learning_rate": 9.904509708309723e-06,
|
| 1592 |
+
"loss": 0.4312,
|
| 1593 |
+
"step": 398
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 23.529411764705884,
|
| 1597 |
+
"grad_norm": 1.0776093006134033,
|
| 1598 |
+
"learning_rate": 9.901828808578846e-06,
|
| 1599 |
+
"loss": 0.4503,
|
| 1600 |
+
"step": 400
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 23.647058823529413,
|
| 1604 |
+
"grad_norm": 1.0322400331497192,
|
| 1605 |
+
"learning_rate": 9.899111167741966e-06,
|
| 1606 |
+
"loss": 0.4685,
|
| 1607 |
+
"step": 402
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 23.764705882352942,
|
| 1611 |
+
"grad_norm": 0.9414217472076416,
|
| 1612 |
+
"learning_rate": 9.896356806168851e-06,
|
| 1613 |
+
"loss": 0.4757,
|
| 1614 |
+
"step": 404
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 23.88235294117647,
|
| 1618 |
+
"grad_norm": 1.1830779314041138,
|
| 1619 |
+
"learning_rate": 9.89356574450451e-06,
|
| 1620 |
+
"loss": 0.4408,
|
| 1621 |
+
"step": 406
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 24.0,
|
| 1625 |
+
"grad_norm": 0.9795882105827332,
|
| 1626 |
+
"learning_rate": 9.890738003669029e-06,
|
| 1627 |
+
"loss": 0.4833,
|
| 1628 |
+
"step": 408
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 24.0,
|
| 1632 |
+
"eval_loss": 0.7643568515777588,
|
| 1633 |
+
"eval_runtime": 37.0004,
|
| 1634 |
+
"eval_samples_per_second": 0.973,
|
| 1635 |
+
"eval_steps_per_second": 0.973,
|
| 1636 |
+
"step": 408
|
| 1637 |
+
},
|
| 1638 |
+
{
|
| 1639 |
+
"epoch": 24.11764705882353,
|
| 1640 |
+
"grad_norm": 1.2952356338500977,
|
| 1641 |
+
"learning_rate": 9.887873604857424e-06,
|
| 1642 |
+
"loss": 0.4299,
|
| 1643 |
+
"step": 410
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 24.235294117647058,
|
| 1647 |
+
"grad_norm": 1.2539509534835815,
|
| 1648 |
+
"learning_rate": 9.884972569539471e-06,
|
| 1649 |
+
"loss": 0.4488,
|
| 1650 |
+
"step": 412
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 24.352941176470587,
|
| 1654 |
+
"grad_norm": 1.0663094520568848,
|
| 1655 |
+
"learning_rate": 9.882034919459556e-06,
|
| 1656 |
+
"loss": 0.4152,
|
| 1657 |
+
"step": 414
|
| 1658 |
+
},
|
| 1659 |
+
{
|
| 1660 |
+
"epoch": 24.470588235294116,
|
| 1661 |
+
"grad_norm": 1.03883957862854,
|
| 1662 |
+
"learning_rate": 9.879060676636502e-06,
|
| 1663 |
+
"loss": 0.4435,
|
| 1664 |
+
"step": 416
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 24.58823529411765,
|
| 1668 |
+
"grad_norm": 0.9475598931312561,
|
| 1669 |
+
"learning_rate": 9.876049863363415e-06,
|
| 1670 |
+
"loss": 0.4408,
|
| 1671 |
+
"step": 418
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 24.705882352941178,
|
| 1675 |
+
"grad_norm": 0.9946249723434448,
|
| 1676 |
+
"learning_rate": 9.873002502207502e-06,
|
| 1677 |
+
"loss": 0.4292,
|
| 1678 |
+
"step": 420
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 24.823529411764707,
|
| 1682 |
+
"grad_norm": 1.135332465171814,
|
| 1683 |
+
"learning_rate": 9.86991861600992e-06,
|
| 1684 |
+
"loss": 0.4057,
|
| 1685 |
+
"step": 422
|
| 1686 |
+
},
|
| 1687 |
+
{
|
| 1688 |
+
"epoch": 24.941176470588236,
|
| 1689 |
+
"grad_norm": 1.7003220319747925,
|
| 1690 |
+
"learning_rate": 9.866798227885588e-06,
|
| 1691 |
+
"loss": 0.4071,
|
| 1692 |
+
"step": 424
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 25.0,
|
| 1696 |
+
"eval_loss": 0.8012258410453796,
|
| 1697 |
+
"eval_runtime": 37.0059,
|
| 1698 |
+
"eval_samples_per_second": 0.973,
|
| 1699 |
+
"eval_steps_per_second": 0.973,
|
| 1700 |
+
"step": 425
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 25.058823529411764,
|
| 1704 |
+
"grad_norm": 1.124939203262329,
|
| 1705 |
+
"learning_rate": 9.863641361223025e-06,
|
| 1706 |
+
"loss": 0.3795,
|
| 1707 |
+
"step": 426
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 25.176470588235293,
|
| 1711 |
+
"grad_norm": 1.2946455478668213,
|
| 1712 |
+
"learning_rate": 9.860448039684169e-06,
|
| 1713 |
+
"loss": 0.3631,
|
| 1714 |
+
"step": 428
|
| 1715 |
+
},
|
| 1716 |
+
{
|
| 1717 |
+
"epoch": 25.294117647058822,
|
| 1718 |
+
"grad_norm": 1.2063255310058594,
|
| 1719 |
+
"learning_rate": 9.857218287204204e-06,
|
| 1720 |
+
"loss": 0.3786,
|
| 1721 |
+
"step": 430
|
| 1722 |
+
},
|
| 1723 |
+
{
|
| 1724 |
+
"epoch": 25.41176470588235,
|
| 1725 |
+
"grad_norm": 1.1833475828170776,
|
| 1726 |
+
"learning_rate": 9.853952127991374e-06,
|
| 1727 |
+
"loss": 0.403,
|
| 1728 |
+
"step": 432
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 25.529411764705884,
|
| 1732 |
+
"grad_norm": 1.2458750009536743,
|
| 1733 |
+
"learning_rate": 9.850649586526808e-06,
|
| 1734 |
+
"loss": 0.3626,
|
| 1735 |
+
"step": 434
|
| 1736 |
+
},
|
| 1737 |
+
{
|
| 1738 |
+
"epoch": 25.647058823529413,
|
| 1739 |
+
"grad_norm": 1.2201582193374634,
|
| 1740 |
+
"learning_rate": 9.847310687564335e-06,
|
| 1741 |
+
"loss": 0.3774,
|
| 1742 |
+
"step": 436
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 25.764705882352942,
|
| 1746 |
+
"grad_norm": 1.2703592777252197,
|
| 1747 |
+
"learning_rate": 9.843935456130295e-06,
|
| 1748 |
+
"loss": 0.4078,
|
| 1749 |
+
"step": 438
|
| 1750 |
+
},
|
| 1751 |
+
{
|
| 1752 |
+
"epoch": 25.88235294117647,
|
| 1753 |
+
"grad_norm": 1.8023614883422852,
|
| 1754 |
+
"learning_rate": 9.840523917523354e-06,
|
| 1755 |
+
"loss": 0.406,
|
| 1756 |
+
"step": 440
|
| 1757 |
+
},
|
| 1758 |
+
{
|
| 1759 |
+
"epoch": 26.0,
|
| 1760 |
+
"grad_norm": 1.3355423212051392,
|
| 1761 |
+
"learning_rate": 9.83707609731432e-06,
|
| 1762 |
+
"loss": 0.3567,
|
| 1763 |
+
"step": 442
|
| 1764 |
+
},
|
| 1765 |
+
{
|
| 1766 |
+
"epoch": 26.0,
|
| 1767 |
+
"eval_loss": 0.8288628458976746,
|
| 1768 |
+
"eval_runtime": 37.0058,
|
| 1769 |
+
"eval_samples_per_second": 0.973,
|
| 1770 |
+
"eval_steps_per_second": 0.973,
|
| 1771 |
+
"step": 442
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 26.0,
|
| 1775 |
+
"step": 442,
|
| 1776 |
+
"total_flos": 3.38458179962667e+17,
|
| 1777 |
+
"train_loss": 0.7497456199173475,
|
| 1778 |
+
"train_runtime": 12081.7026,
|
| 1779 |
+
"train_samples_per_second": 1.689,
|
| 1780 |
+
"train_steps_per_second": 0.211
|
| 1781 |
+
}
|
| 1782 |
+
],
|
| 1783 |
+
"logging_steps": 2,
|
| 1784 |
+
"max_steps": 2550,
|
| 1785 |
+
"num_input_tokens_seen": 0,
|
| 1786 |
+
"num_train_epochs": 150,
|
| 1787 |
+
"save_steps": 25,
|
| 1788 |
+
"stateful_callbacks": {
|
| 1789 |
+
"EarlyStoppingCallback": {
|
| 1790 |
+
"args": {
|
| 1791 |
+
"early_stopping_patience": 7,
|
| 1792 |
+
"early_stopping_threshold": 0.0
|
| 1793 |
+
},
|
| 1794 |
+
"attributes": {
|
| 1795 |
+
"early_stopping_patience_counter": 0
|
| 1796 |
+
}
|
| 1797 |
+
},
|
| 1798 |
+
"TrainerControl": {
|
| 1799 |
+
"args": {
|
| 1800 |
+
"should_epoch_stop": false,
|
| 1801 |
+
"should_evaluate": false,
|
| 1802 |
+
"should_log": false,
|
| 1803 |
+
"should_save": true,
|
| 1804 |
+
"should_training_stop": true
|
| 1805 |
+
},
|
| 1806 |
+
"attributes": {}
|
| 1807 |
+
}
|
| 1808 |
+
},
|
| 1809 |
+
"total_flos": 3.38458179962667e+17,
|
| 1810 |
+
"train_batch_size": 1,
|
| 1811 |
+
"trial_name": null,
|
| 1812 |
+
"trial_params": null
|
| 1813 |
+
}
|