Commit
•
fa3758e
1
Parent(s):
ebd5aa4
Model save
Browse files- README.md +105 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +1170 -0
README.md
ADDED
|
@@ -0,0 +1,105 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
| 3 |
+
library_name: peft
|
| 4 |
+
license: llama3.1
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: Llama-31-8B_task-2_120-samples_config-4
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# Llama-31-8B_task-2_120-samples_config-4
|
| 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: 0.7395
|
| 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: 16
|
| 46 |
+
- total_train_batch_size: 16
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 50 |
+
- num_epochs: 150
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 55 |
+
|:-------------:|:-------:|:----:|:---------------:|
|
| 56 |
+
| 1.1544 | 0.9091 | 5 | 1.1237 |
|
| 57 |
+
| 1.132 | 2.0 | 11 | 1.1193 |
|
| 58 |
+
| 1.0689 | 2.9091 | 16 | 1.1136 |
|
| 59 |
+
| 1.0956 | 4.0 | 22 | 1.1011 |
|
| 60 |
+
| 1.1157 | 4.9091 | 27 | 1.0871 |
|
| 61 |
+
| 1.0778 | 6.0 | 33 | 1.0639 |
|
| 62 |
+
| 1.0458 | 6.9091 | 38 | 1.0393 |
|
| 63 |
+
| 0.9854 | 8.0 | 44 | 1.0027 |
|
| 64 |
+
| 0.9996 | 8.9091 | 49 | 0.9696 |
|
| 65 |
+
| 0.8991 | 10.0 | 55 | 0.9317 |
|
| 66 |
+
| 0.8897 | 10.9091 | 60 | 0.9052 |
|
| 67 |
+
| 0.8711 | 12.0 | 66 | 0.8788 |
|
| 68 |
+
| 0.8809 | 12.9091 | 71 | 0.8588 |
|
| 69 |
+
| 0.7972 | 14.0 | 77 | 0.8368 |
|
| 70 |
+
| 0.8156 | 14.9091 | 82 | 0.8208 |
|
| 71 |
+
| 0.7815 | 16.0 | 88 | 0.8057 |
|
| 72 |
+
| 0.7492 | 16.9091 | 93 | 0.7956 |
|
| 73 |
+
| 0.7587 | 18.0 | 99 | 0.7855 |
|
| 74 |
+
| 0.7483 | 18.9091 | 104 | 0.7780 |
|
| 75 |
+
| 0.7296 | 20.0 | 110 | 0.7695 |
|
| 76 |
+
| 0.7441 | 20.9091 | 115 | 0.7629 |
|
| 77 |
+
| 0.7176 | 22.0 | 121 | 0.7561 |
|
| 78 |
+
| 0.7033 | 22.9091 | 126 | 0.7508 |
|
| 79 |
+
| 0.6906 | 24.0 | 132 | 0.7443 |
|
| 80 |
+
| 0.6954 | 24.9091 | 137 | 0.7396 |
|
| 81 |
+
| 0.6578 | 26.0 | 143 | 0.7344 |
|
| 82 |
+
| 0.6495 | 26.9091 | 148 | 0.7310 |
|
| 83 |
+
| 0.6391 | 28.0 | 154 | 0.7269 |
|
| 84 |
+
| 0.6442 | 28.9091 | 159 | 0.7237 |
|
| 85 |
+
| 0.6268 | 30.0 | 165 | 0.7199 |
|
| 86 |
+
| 0.6536 | 30.9091 | 170 | 0.7183 |
|
| 87 |
+
| 0.6092 | 32.0 | 176 | 0.7163 |
|
| 88 |
+
| 0.621 | 32.9091 | 181 | 0.7149 |
|
| 89 |
+
| 0.5823 | 34.0 | 187 | 0.7144 |
|
| 90 |
+
| 0.5651 | 34.9091 | 192 | 0.7156 |
|
| 91 |
+
| 0.5951 | 36.0 | 198 | 0.7164 |
|
| 92 |
+
| 0.5637 | 36.9091 | 203 | 0.7195 |
|
| 93 |
+
| 0.5669 | 38.0 | 209 | 0.7219 |
|
| 94 |
+
| 0.5613 | 38.9091 | 214 | 0.7278 |
|
| 95 |
+
| 0.5156 | 40.0 | 220 | 0.7309 |
|
| 96 |
+
| 0.5044 | 40.9091 | 225 | 0.7395 |
|
| 97 |
+
|
| 98 |
+
|
| 99 |
+
### Framework versions
|
| 100 |
+
|
| 101 |
+
- PEFT 0.12.0
|
| 102 |
+
- Transformers 4.44.0
|
| 103 |
+
- Pytorch 2.1.2+cu121
|
| 104 |
+
- Datasets 2.20.0
|
| 105 |
+
- 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 sha256:8299812670991a96d797eddf296214e83323686aec3515d2d9287e07c0fb0dae
|
| 3 |
size 167832240
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.90909090909091,
|
| 3 |
+
"total_flos": 3.429455470966866e+17,
|
| 4 |
+
"train_loss": 0.7736989471647474,
|
| 5 |
+
"train_runtime": 12307.7975,
|
| 6 |
+
"train_samples": 88,
|
| 7 |
+
"train_samples_per_second": 1.072,
|
| 8 |
+
"train_steps_per_second": 0.061
|
| 9 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.90909090909091,
|
| 3 |
+
"total_flos": 3.429455470966866e+17,
|
| 4 |
+
"train_loss": 0.7736989471647474,
|
| 5 |
+
"train_runtime": 12307.7975,
|
| 6 |
+
"train_samples": 88,
|
| 7 |
+
"train_samples_per_second": 1.072,
|
| 8 |
+
"train_steps_per_second": 0.061
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1170 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7143663763999939,
|
| 3 |
+
"best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-4/checkpoint-187",
|
| 4 |
+
"epoch": 40.90909090909091,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 225,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.18181818181818182,
|
| 13 |
+
"grad_norm": 0.7671908140182495,
|
| 14 |
+
"learning_rate": 1.3333333333333336e-07,
|
| 15 |
+
"loss": 1.1716,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.36363636363636365,
|
| 20 |
+
"grad_norm": 0.6510481834411621,
|
| 21 |
+
"learning_rate": 2.666666666666667e-07,
|
| 22 |
+
"loss": 1.0331,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.7272727272727273,
|
| 27 |
+
"grad_norm": 0.702520489692688,
|
| 28 |
+
"learning_rate": 5.333333333333335e-07,
|
| 29 |
+
"loss": 1.1544,
|
| 30 |
+
"step": 4
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.9090909090909091,
|
| 34 |
+
"eval_loss": 1.1236884593963623,
|
| 35 |
+
"eval_runtime": 24.3567,
|
| 36 |
+
"eval_samples_per_second": 0.985,
|
| 37 |
+
"eval_steps_per_second": 0.985,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0909090909090908,
|
| 42 |
+
"grad_norm": 0.7136379480361938,
|
| 43 |
+
"learning_rate": 8.000000000000001e-07,
|
| 44 |
+
"loss": 1.0735,
|
| 45 |
+
"step": 6
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.4545454545454546,
|
| 49 |
+
"grad_norm": 0.7056083083152771,
|
| 50 |
+
"learning_rate": 1.066666666666667e-06,
|
| 51 |
+
"loss": 1.1336,
|
| 52 |
+
"step": 8
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.8181818181818183,
|
| 56 |
+
"grad_norm": 0.7233437895774841,
|
| 57 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 58 |
+
"loss": 1.132,
|
| 59 |
+
"step": 10
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 2.0,
|
| 63 |
+
"eval_loss": 1.1193286180496216,
|
| 64 |
+
"eval_runtime": 24.3549,
|
| 65 |
+
"eval_samples_per_second": 0.985,
|
| 66 |
+
"eval_steps_per_second": 0.985,
|
| 67 |
+
"step": 11
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.1818181818181817,
|
| 71 |
+
"grad_norm": 0.6547144055366516,
|
| 72 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 73 |
+
"loss": 1.0473,
|
| 74 |
+
"step": 12
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.5454545454545454,
|
| 78 |
+
"grad_norm": 0.7151400446891785,
|
| 79 |
+
"learning_rate": 1.8666666666666669e-06,
|
| 80 |
+
"loss": 1.1478,
|
| 81 |
+
"step": 14
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.909090909090909,
|
| 85 |
+
"grad_norm": 0.7789710760116577,
|
| 86 |
+
"learning_rate": 2.133333333333334e-06,
|
| 87 |
+
"loss": 1.0689,
|
| 88 |
+
"step": 16
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 2.909090909090909,
|
| 92 |
+
"eval_loss": 1.1136311292648315,
|
| 93 |
+
"eval_runtime": 24.3553,
|
| 94 |
+
"eval_samples_per_second": 0.985,
|
| 95 |
+
"eval_steps_per_second": 0.985,
|
| 96 |
+
"step": 16
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.2727272727272725,
|
| 100 |
+
"grad_norm": 0.6943601369857788,
|
| 101 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 102 |
+
"loss": 1.1271,
|
| 103 |
+
"step": 18
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 3.6363636363636362,
|
| 107 |
+
"grad_norm": 0.691636860370636,
|
| 108 |
+
"learning_rate": 2.666666666666667e-06,
|
| 109 |
+
"loss": 1.0717,
|
| 110 |
+
"step": 20
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 4.0,
|
| 114 |
+
"grad_norm": 0.6316518187522888,
|
| 115 |
+
"learning_rate": 2.9333333333333338e-06,
|
| 116 |
+
"loss": 1.0956,
|
| 117 |
+
"step": 22
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 4.0,
|
| 121 |
+
"eval_loss": 1.1010679006576538,
|
| 122 |
+
"eval_runtime": 24.358,
|
| 123 |
+
"eval_samples_per_second": 0.985,
|
| 124 |
+
"eval_steps_per_second": 0.985,
|
| 125 |
+
"step": 22
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 4.363636363636363,
|
| 129 |
+
"grad_norm": 0.7023854851722717,
|
| 130 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 131 |
+
"loss": 1.1054,
|
| 132 |
+
"step": 24
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 4.7272727272727275,
|
| 136 |
+
"grad_norm": 0.7613805532455444,
|
| 137 |
+
"learning_rate": 3.4666666666666672e-06,
|
| 138 |
+
"loss": 1.1157,
|
| 139 |
+
"step": 26
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 4.909090909090909,
|
| 143 |
+
"eval_loss": 1.087128758430481,
|
| 144 |
+
"eval_runtime": 24.3561,
|
| 145 |
+
"eval_samples_per_second": 0.985,
|
| 146 |
+
"eval_steps_per_second": 0.985,
|
| 147 |
+
"step": 27
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.090909090909091,
|
| 151 |
+
"grad_norm": 0.6336562633514404,
|
| 152 |
+
"learning_rate": 3.7333333333333337e-06,
|
| 153 |
+
"loss": 0.9922,
|
| 154 |
+
"step": 28
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.454545454545454,
|
| 158 |
+
"grad_norm": 0.7272669076919556,
|
| 159 |
+
"learning_rate": 4.000000000000001e-06,
|
| 160 |
+
"loss": 1.0947,
|
| 161 |
+
"step": 30
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.818181818181818,
|
| 165 |
+
"grad_norm": 0.7287123203277588,
|
| 166 |
+
"learning_rate": 4.266666666666668e-06,
|
| 167 |
+
"loss": 1.0778,
|
| 168 |
+
"step": 32
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 6.0,
|
| 172 |
+
"eval_loss": 1.063869833946228,
|
| 173 |
+
"eval_runtime": 24.3649,
|
| 174 |
+
"eval_samples_per_second": 0.985,
|
| 175 |
+
"eval_steps_per_second": 0.985,
|
| 176 |
+
"step": 33
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.181818181818182,
|
| 180 |
+
"grad_norm": 0.6734246015548706,
|
| 181 |
+
"learning_rate": 4.533333333333334e-06,
|
| 182 |
+
"loss": 1.038,
|
| 183 |
+
"step": 34
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 6.545454545454545,
|
| 187 |
+
"grad_norm": 0.628820538520813,
|
| 188 |
+
"learning_rate": 4.800000000000001e-06,
|
| 189 |
+
"loss": 1.03,
|
| 190 |
+
"step": 36
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 6.909090909090909,
|
| 194 |
+
"grad_norm": 0.6829433441162109,
|
| 195 |
+
"learning_rate": 5.0666666666666676e-06,
|
| 196 |
+
"loss": 1.0458,
|
| 197 |
+
"step": 38
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 6.909090909090909,
|
| 201 |
+
"eval_loss": 1.039298176765442,
|
| 202 |
+
"eval_runtime": 24.363,
|
| 203 |
+
"eval_samples_per_second": 0.985,
|
| 204 |
+
"eval_steps_per_second": 0.985,
|
| 205 |
+
"step": 38
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 7.2727272727272725,
|
| 209 |
+
"grad_norm": 0.684455931186676,
|
| 210 |
+
"learning_rate": 5.333333333333334e-06,
|
| 211 |
+
"loss": 1.0333,
|
| 212 |
+
"step": 40
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 7.636363636363637,
|
| 216 |
+
"grad_norm": 0.6354673504829407,
|
| 217 |
+
"learning_rate": 5.600000000000001e-06,
|
| 218 |
+
"loss": 0.9913,
|
| 219 |
+
"step": 42
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 8.0,
|
| 223 |
+
"grad_norm": 0.5571680665016174,
|
| 224 |
+
"learning_rate": 5.8666666666666675e-06,
|
| 225 |
+
"loss": 0.9854,
|
| 226 |
+
"step": 44
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 8.0,
|
| 230 |
+
"eval_loss": 1.0027111768722534,
|
| 231 |
+
"eval_runtime": 24.3512,
|
| 232 |
+
"eval_samples_per_second": 0.986,
|
| 233 |
+
"eval_steps_per_second": 0.986,
|
| 234 |
+
"step": 44
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 8.363636363636363,
|
| 238 |
+
"grad_norm": 0.5560793876647949,
|
| 239 |
+
"learning_rate": 6.133333333333334e-06,
|
| 240 |
+
"loss": 0.9371,
|
| 241 |
+
"step": 46
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 8.727272727272727,
|
| 245 |
+
"grad_norm": 0.6101187467575073,
|
| 246 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 247 |
+
"loss": 0.9996,
|
| 248 |
+
"step": 48
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 8.909090909090908,
|
| 252 |
+
"eval_loss": 0.9696131348609924,
|
| 253 |
+
"eval_runtime": 24.3608,
|
| 254 |
+
"eval_samples_per_second": 0.985,
|
| 255 |
+
"eval_steps_per_second": 0.985,
|
| 256 |
+
"step": 49
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 9.090909090909092,
|
| 260 |
+
"grad_norm": 0.5255017280578613,
|
| 261 |
+
"learning_rate": 6.666666666666667e-06,
|
| 262 |
+
"loss": 0.9569,
|
| 263 |
+
"step": 50
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.454545454545455,
|
| 267 |
+
"grad_norm": 0.5077732801437378,
|
| 268 |
+
"learning_rate": 6.9333333333333344e-06,
|
| 269 |
+
"loss": 0.9602,
|
| 270 |
+
"step": 52
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.818181818181818,
|
| 274 |
+
"grad_norm": 0.4885198771953583,
|
| 275 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 276 |
+
"loss": 0.8991,
|
| 277 |
+
"step": 54
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 10.0,
|
| 281 |
+
"eval_loss": 0.9317263960838318,
|
| 282 |
+
"eval_runtime": 24.3522,
|
| 283 |
+
"eval_samples_per_second": 0.986,
|
| 284 |
+
"eval_steps_per_second": 0.986,
|
| 285 |
+
"step": 55
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 10.181818181818182,
|
| 289 |
+
"grad_norm": 0.42795872688293457,
|
| 290 |
+
"learning_rate": 7.4666666666666675e-06,
|
| 291 |
+
"loss": 0.9282,
|
| 292 |
+
"step": 56
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 10.545454545454545,
|
| 296 |
+
"grad_norm": 0.3717321455478668,
|
| 297 |
+
"learning_rate": 7.733333333333334e-06,
|
| 298 |
+
"loss": 0.9173,
|
| 299 |
+
"step": 58
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 10.909090909090908,
|
| 303 |
+
"grad_norm": 0.37956222891807556,
|
| 304 |
+
"learning_rate": 8.000000000000001e-06,
|
| 305 |
+
"loss": 0.8897,
|
| 306 |
+
"step": 60
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 10.909090909090908,
|
| 310 |
+
"eval_loss": 0.9051777720451355,
|
| 311 |
+
"eval_runtime": 24.3554,
|
| 312 |
+
"eval_samples_per_second": 0.985,
|
| 313 |
+
"eval_steps_per_second": 0.985,
|
| 314 |
+
"step": 60
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 11.272727272727273,
|
| 318 |
+
"grad_norm": 0.32997962832450867,
|
| 319 |
+
"learning_rate": 8.266666666666667e-06,
|
| 320 |
+
"loss": 0.8511,
|
| 321 |
+
"step": 62
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 11.636363636363637,
|
| 325 |
+
"grad_norm": 0.3067737817764282,
|
| 326 |
+
"learning_rate": 8.533333333333335e-06,
|
| 327 |
+
"loss": 0.8714,
|
| 328 |
+
"step": 64
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 12.0,
|
| 332 |
+
"grad_norm": 0.3460928797721863,
|
| 333 |
+
"learning_rate": 8.8e-06,
|
| 334 |
+
"loss": 0.8711,
|
| 335 |
+
"step": 66
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 12.0,
|
| 339 |
+
"eval_loss": 0.878821074962616,
|
| 340 |
+
"eval_runtime": 24.3541,
|
| 341 |
+
"eval_samples_per_second": 0.985,
|
| 342 |
+
"eval_steps_per_second": 0.985,
|
| 343 |
+
"step": 66
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 12.363636363636363,
|
| 347 |
+
"grad_norm": 0.29182004928588867,
|
| 348 |
+
"learning_rate": 9.066666666666667e-06,
|
| 349 |
+
"loss": 0.8274,
|
| 350 |
+
"step": 68
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.727272727272727,
|
| 354 |
+
"grad_norm": 0.31693214178085327,
|
| 355 |
+
"learning_rate": 9.333333333333334e-06,
|
| 356 |
+
"loss": 0.8809,
|
| 357 |
+
"step": 70
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 12.909090909090908,
|
| 361 |
+
"eval_loss": 0.8587873578071594,
|
| 362 |
+
"eval_runtime": 24.3566,
|
| 363 |
+
"eval_samples_per_second": 0.985,
|
| 364 |
+
"eval_steps_per_second": 0.985,
|
| 365 |
+
"step": 71
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 13.090909090909092,
|
| 369 |
+
"grad_norm": 0.29670482873916626,
|
| 370 |
+
"learning_rate": 9.600000000000001e-06,
|
| 371 |
+
"loss": 0.8263,
|
| 372 |
+
"step": 72
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 13.454545454545455,
|
| 376 |
+
"grad_norm": 0.2604750096797943,
|
| 377 |
+
"learning_rate": 9.866666666666668e-06,
|
| 378 |
+
"loss": 0.8237,
|
| 379 |
+
"step": 74
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 13.818181818181818,
|
| 383 |
+
"grad_norm": 0.2680836021900177,
|
| 384 |
+
"learning_rate": 9.999945845889795e-06,
|
| 385 |
+
"loss": 0.7972,
|
| 386 |
+
"step": 76
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 14.0,
|
| 390 |
+
"eval_loss": 0.8367733955383301,
|
| 391 |
+
"eval_runtime": 24.3654,
|
| 392 |
+
"eval_samples_per_second": 0.985,
|
| 393 |
+
"eval_steps_per_second": 0.985,
|
| 394 |
+
"step": 77
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 14.181818181818182,
|
| 398 |
+
"grad_norm": 0.2775014638900757,
|
| 399 |
+
"learning_rate": 9.999512620046523e-06,
|
| 400 |
+
"loss": 0.8129,
|
| 401 |
+
"step": 78
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 14.545454545454545,
|
| 405 |
+
"grad_norm": 0.24960099160671234,
|
| 406 |
+
"learning_rate": 9.99864620589731e-06,
|
| 407 |
+
"loss": 0.8163,
|
| 408 |
+
"step": 80
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 14.909090909090908,
|
| 412 |
+
"grad_norm": 0.2805323004722595,
|
| 413 |
+
"learning_rate": 9.99734667851357e-06,
|
| 414 |
+
"loss": 0.8156,
|
| 415 |
+
"step": 82
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 14.909090909090908,
|
| 419 |
+
"eval_loss": 0.8207753300666809,
|
| 420 |
+
"eval_runtime": 24.3571,
|
| 421 |
+
"eval_samples_per_second": 0.985,
|
| 422 |
+
"eval_steps_per_second": 0.985,
|
| 423 |
+
"step": 82
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 15.272727272727273,
|
| 427 |
+
"grad_norm": 0.23699168860912323,
|
| 428 |
+
"learning_rate": 9.995614150494293e-06,
|
| 429 |
+
"loss": 0.7658,
|
| 430 |
+
"step": 84
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 15.636363636363637,
|
| 434 |
+
"grad_norm": 0.24089692533016205,
|
| 435 |
+
"learning_rate": 9.993448771956285e-06,
|
| 436 |
+
"loss": 0.8111,
|
| 437 |
+
"step": 86
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 16.0,
|
| 441 |
+
"grad_norm": 0.24843314290046692,
|
| 442 |
+
"learning_rate": 9.99085073052117e-06,
|
| 443 |
+
"loss": 0.7815,
|
| 444 |
+
"step": 88
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 16.0,
|
| 448 |
+
"eval_loss": 0.8057171702384949,
|
| 449 |
+
"eval_runtime": 24.3555,
|
| 450 |
+
"eval_samples_per_second": 0.985,
|
| 451 |
+
"eval_steps_per_second": 0.985,
|
| 452 |
+
"step": 88
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 16.363636363636363,
|
| 456 |
+
"grad_norm": 0.2455184906721115,
|
| 457 |
+
"learning_rate": 9.987820251299121e-06,
|
| 458 |
+
"loss": 0.7832,
|
| 459 |
+
"step": 90
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 16.727272727272727,
|
| 463 |
+
"grad_norm": 0.23427604138851166,
|
| 464 |
+
"learning_rate": 9.984357596869369e-06,
|
| 465 |
+
"loss": 0.7492,
|
| 466 |
+
"step": 92
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 16.90909090909091,
|
| 470 |
+
"eval_loss": 0.7956312298774719,
|
| 471 |
+
"eval_runtime": 24.3597,
|
| 472 |
+
"eval_samples_per_second": 0.985,
|
| 473 |
+
"eval_steps_per_second": 0.985,
|
| 474 |
+
"step": 93
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 17.09090909090909,
|
| 478 |
+
"grad_norm": 0.2534084618091583,
|
| 479 |
+
"learning_rate": 9.980463067257437e-06,
|
| 480 |
+
"loss": 0.774,
|
| 481 |
+
"step": 94
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 17.454545454545453,
|
| 485 |
+
"grad_norm": 0.2455044388771057,
|
| 486 |
+
"learning_rate": 9.976136999909156e-06,
|
| 487 |
+
"loss": 0.7431,
|
| 488 |
+
"step": 96
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 17.818181818181817,
|
| 492 |
+
"grad_norm": 0.22460192441940308,
|
| 493 |
+
"learning_rate": 9.971379769661422e-06,
|
| 494 |
+
"loss": 0.7587,
|
| 495 |
+
"step": 98
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 18.0,
|
| 499 |
+
"eval_loss": 0.7855010628700256,
|
| 500 |
+
"eval_runtime": 24.3575,
|
| 501 |
+
"eval_samples_per_second": 0.985,
|
| 502 |
+
"eval_steps_per_second": 0.985,
|
| 503 |
+
"step": 99
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 18.181818181818183,
|
| 507 |
+
"grad_norm": 0.2319430559873581,
|
| 508 |
+
"learning_rate": 9.966191788709716e-06,
|
| 509 |
+
"loss": 0.7886,
|
| 510 |
+
"step": 100
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 18.545454545454547,
|
| 514 |
+
"grad_norm": 0.22566601634025574,
|
| 515 |
+
"learning_rate": 9.960573506572391e-06,
|
| 516 |
+
"loss": 0.7468,
|
| 517 |
+
"step": 102
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 18.90909090909091,
|
| 521 |
+
"grad_norm": 0.2197207510471344,
|
| 522 |
+
"learning_rate": 9.95452541005172e-06,
|
| 523 |
+
"loss": 0.7483,
|
| 524 |
+
"step": 104
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 18.90909090909091,
|
| 528 |
+
"eval_loss": 0.7779585719108582,
|
| 529 |
+
"eval_runtime": 24.351,
|
| 530 |
+
"eval_samples_per_second": 0.986,
|
| 531 |
+
"eval_steps_per_second": 0.986,
|
| 532 |
+
"step": 104
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 19.272727272727273,
|
| 536 |
+
"grad_norm": 0.25423386693000793,
|
| 537 |
+
"learning_rate": 9.948048023191728e-06,
|
| 538 |
+
"loss": 0.7066,
|
| 539 |
+
"step": 106
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 19.636363636363637,
|
| 543 |
+
"grad_norm": 0.27388209104537964,
|
| 544 |
+
"learning_rate": 9.941141907232766e-06,
|
| 545 |
+
"loss": 0.7644,
|
| 546 |
+
"step": 108
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 20.0,
|
| 550 |
+
"grad_norm": 0.21174727380275726,
|
| 551 |
+
"learning_rate": 9.933807660562898e-06,
|
| 552 |
+
"loss": 0.7296,
|
| 553 |
+
"step": 110
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 20.0,
|
| 557 |
+
"eval_loss": 0.7694642543792725,
|
| 558 |
+
"eval_runtime": 24.3585,
|
| 559 |
+
"eval_samples_per_second": 0.985,
|
| 560 |
+
"eval_steps_per_second": 0.985,
|
| 561 |
+
"step": 110
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 20.363636363636363,
|
| 565 |
+
"grad_norm": 0.21564796566963196,
|
| 566 |
+
"learning_rate": 9.926045918666045e-06,
|
| 567 |
+
"loss": 0.7005,
|
| 568 |
+
"step": 112
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 20.727272727272727,
|
| 572 |
+
"grad_norm": 0.21516019105911255,
|
| 573 |
+
"learning_rate": 9.91785735406693e-06,
|
| 574 |
+
"loss": 0.7441,
|
| 575 |
+
"step": 114
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 20.90909090909091,
|
| 579 |
+
"eval_loss": 0.7629389762878418,
|
| 580 |
+
"eval_runtime": 24.3562,
|
| 581 |
+
"eval_samples_per_second": 0.985,
|
| 582 |
+
"eval_steps_per_second": 0.985,
|
| 583 |
+
"step": 115
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 21.09090909090909,
|
| 587 |
+
"grad_norm": 0.25928786396980286,
|
| 588 |
+
"learning_rate": 9.909242676272797e-06,
|
| 589 |
+
"loss": 0.7448,
|
| 590 |
+
"step": 116
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 21.454545454545453,
|
| 594 |
+
"grad_norm": 0.22024983167648315,
|
| 595 |
+
"learning_rate": 9.90020263171194e-06,
|
| 596 |
+
"loss": 0.7277,
|
| 597 |
+
"step": 118
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 21.818181818181817,
|
| 601 |
+
"grad_norm": 0.22692517936229706,
|
| 602 |
+
"learning_rate": 9.890738003669029e-06,
|
| 603 |
+
"loss": 0.7176,
|
| 604 |
+
"step": 120
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 22.0,
|
| 608 |
+
"eval_loss": 0.7560660243034363,
|
| 609 |
+
"eval_runtime": 24.3539,
|
| 610 |
+
"eval_samples_per_second": 0.985,
|
| 611 |
+
"eval_steps_per_second": 0.985,
|
| 612 |
+
"step": 121
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 22.181818181818183,
|
| 616 |
+
"grad_norm": 0.2096530646085739,
|
| 617 |
+
"learning_rate": 9.880849612217238e-06,
|
| 618 |
+
"loss": 0.6737,
|
| 619 |
+
"step": 122
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 22.545454545454547,
|
| 623 |
+
"grad_norm": 0.23917944729328156,
|
| 624 |
+
"learning_rate": 9.870538314147194e-06,
|
| 625 |
+
"loss": 0.7373,
|
| 626 |
+
"step": 124
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 22.90909090909091,
|
| 630 |
+
"grad_norm": 0.2294284701347351,
|
| 631 |
+
"learning_rate": 9.859805002892733e-06,
|
| 632 |
+
"loss": 0.7033,
|
| 633 |
+
"step": 126
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 22.90909090909091,
|
| 637 |
+
"eval_loss": 0.7507641911506653,
|
| 638 |
+
"eval_runtime": 24.3597,
|
| 639 |
+
"eval_samples_per_second": 0.985,
|
| 640 |
+
"eval_steps_per_second": 0.985,
|
| 641 |
+
"step": 126
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 23.272727272727273,
|
| 645 |
+
"grad_norm": 0.22704288363456726,
|
| 646 |
+
"learning_rate": 9.84865060845349e-06,
|
| 647 |
+
"loss": 0.6932,
|
| 648 |
+
"step": 128
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 23.636363636363637,
|
| 652 |
+
"grad_norm": 0.2241230458021164,
|
| 653 |
+
"learning_rate": 9.83707609731432e-06,
|
| 654 |
+
"loss": 0.6907,
|
| 655 |
+
"step": 130
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 24.0,
|
| 659 |
+
"grad_norm": 0.2692793011665344,
|
| 660 |
+
"learning_rate": 9.825082472361558e-06,
|
| 661 |
+
"loss": 0.6906,
|
| 662 |
+
"step": 132
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 24.0,
|
| 666 |
+
"eval_loss": 0.7443225383758545,
|
| 667 |
+
"eval_runtime": 24.3494,
|
| 668 |
+
"eval_samples_per_second": 0.986,
|
| 669 |
+
"eval_steps_per_second": 0.986,
|
| 670 |
+
"step": 132
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 24.363636363636363,
|
| 674 |
+
"grad_norm": 0.2639993727207184,
|
| 675 |
+
"learning_rate": 9.812670772796113e-06,
|
| 676 |
+
"loss": 0.7003,
|
| 677 |
+
"step": 134
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 24.727272727272727,
|
| 681 |
+
"grad_norm": 0.2364564836025238,
|
| 682 |
+
"learning_rate": 9.799842074043438e-06,
|
| 683 |
+
"loss": 0.6954,
|
| 684 |
+
"step": 136
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 24.90909090909091,
|
| 688 |
+
"eval_loss": 0.7396196722984314,
|
| 689 |
+
"eval_runtime": 24.3588,
|
| 690 |
+
"eval_samples_per_second": 0.985,
|
| 691 |
+
"eval_steps_per_second": 0.985,
|
| 692 |
+
"step": 137
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 25.09090909090909,
|
| 696 |
+
"grad_norm": 0.25801482796669006,
|
| 697 |
+
"learning_rate": 9.786597487660336e-06,
|
| 698 |
+
"loss": 0.6571,
|
| 699 |
+
"step": 138
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 25.454545454545453,
|
| 703 |
+
"grad_norm": 0.23688581585884094,
|
| 704 |
+
"learning_rate": 9.77293816123866e-06,
|
| 705 |
+
"loss": 0.6904,
|
| 706 |
+
"step": 140
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 25.818181818181817,
|
| 710 |
+
"grad_norm": 0.2453926056623459,
|
| 711 |
+
"learning_rate": 9.75886527830587e-06,
|
| 712 |
+
"loss": 0.6578,
|
| 713 |
+
"step": 142
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 26.0,
|
| 717 |
+
"eval_loss": 0.734368622303009,
|
| 718 |
+
"eval_runtime": 24.3632,
|
| 719 |
+
"eval_samples_per_second": 0.985,
|
| 720 |
+
"eval_steps_per_second": 0.985,
|
| 721 |
+
"step": 143
|
| 722 |
+
},
|
| 723 |
+
{
|
| 724 |
+
"epoch": 26.181818181818183,
|
| 725 |
+
"grad_norm": 0.2570464611053467,
|
| 726 |
+
"learning_rate": 9.744380058222483e-06,
|
| 727 |
+
"loss": 0.6708,
|
| 728 |
+
"step": 144
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 26.545454545454547,
|
| 732 |
+
"grad_norm": 0.2480405867099762,
|
| 733 |
+
"learning_rate": 9.729483756076436e-06,
|
| 734 |
+
"loss": 0.6975,
|
| 735 |
+
"step": 146
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 26.90909090909091,
|
| 739 |
+
"grad_norm": 0.2474363148212433,
|
| 740 |
+
"learning_rate": 9.714177662574316e-06,
|
| 741 |
+
"loss": 0.6495,
|
| 742 |
+
"step": 148
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 26.90909090909091,
|
| 746 |
+
"eval_loss": 0.7309517860412598,
|
| 747 |
+
"eval_runtime": 24.351,
|
| 748 |
+
"eval_samples_per_second": 0.986,
|
| 749 |
+
"eval_steps_per_second": 0.986,
|
| 750 |
+
"step": 148
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 27.272727272727273,
|
| 754 |
+
"grad_norm": 0.2587747871875763,
|
| 755 |
+
"learning_rate": 9.698463103929542e-06,
|
| 756 |
+
"loss": 0.6405,
|
| 757 |
+
"step": 150
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 27.636363636363637,
|
| 761 |
+
"grad_norm": 0.2785438597202301,
|
| 762 |
+
"learning_rate": 9.682341441747446e-06,
|
| 763 |
+
"loss": 0.6908,
|
| 764 |
+
"step": 152
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 28.0,
|
| 768 |
+
"grad_norm": 0.25493577122688293,
|
| 769 |
+
"learning_rate": 9.665814072907293e-06,
|
| 770 |
+
"loss": 0.6391,
|
| 771 |
+
"step": 154
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 28.0,
|
| 775 |
+
"eval_loss": 0.7269113063812256,
|
| 776 |
+
"eval_runtime": 24.36,
|
| 777 |
+
"eval_samples_per_second": 0.985,
|
| 778 |
+
"eval_steps_per_second": 0.985,
|
| 779 |
+
"step": 154
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 28.363636363636363,
|
| 783 |
+
"grad_norm": 0.25702783465385437,
|
| 784 |
+
"learning_rate": 9.648882429441258e-06,
|
| 785 |
+
"loss": 0.6446,
|
| 786 |
+
"step": 156
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 28.727272727272727,
|
| 790 |
+
"grad_norm": 0.2598085105419159,
|
| 791 |
+
"learning_rate": 9.63154797841033e-06,
|
| 792 |
+
"loss": 0.6442,
|
| 793 |
+
"step": 158
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 28.90909090909091,
|
| 797 |
+
"eval_loss": 0.723728358745575,
|
| 798 |
+
"eval_runtime": 24.3516,
|
| 799 |
+
"eval_samples_per_second": 0.986,
|
| 800 |
+
"eval_steps_per_second": 0.986,
|
| 801 |
+
"step": 159
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 29.09090909090909,
|
| 805 |
+
"grad_norm": 0.27804017066955566,
|
| 806 |
+
"learning_rate": 9.613812221777212e-06,
|
| 807 |
+
"loss": 0.6514,
|
| 808 |
+
"step": 160
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 29.454545454545453,
|
| 812 |
+
"grad_norm": 0.2686798572540283,
|
| 813 |
+
"learning_rate": 9.595676696276173e-06,
|
| 814 |
+
"loss": 0.6613,
|
| 815 |
+
"step": 162
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 29.818181818181817,
|
| 819 |
+
"grad_norm": 0.275974303483963,
|
| 820 |
+
"learning_rate": 9.577142973279896e-06,
|
| 821 |
+
"loss": 0.6268,
|
| 822 |
+
"step": 164
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 30.0,
|
| 826 |
+
"eval_loss": 0.7199351787567139,
|
| 827 |
+
"eval_runtime": 24.3608,
|
| 828 |
+
"eval_samples_per_second": 0.985,
|
| 829 |
+
"eval_steps_per_second": 0.985,
|
| 830 |
+
"step": 165
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 30.181818181818183,
|
| 834 |
+
"grad_norm": 0.2688060700893402,
|
| 835 |
+
"learning_rate": 9.55821265866333e-06,
|
| 836 |
+
"loss": 0.6002,
|
| 837 |
+
"step": 166
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 30.545454545454547,
|
| 841 |
+
"grad_norm": 0.27310389280319214,
|
| 842 |
+
"learning_rate": 9.538887392664544e-06,
|
| 843 |
+
"loss": 0.6103,
|
| 844 |
+
"step": 168
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 30.90909090909091,
|
| 848 |
+
"grad_norm": 0.294357031583786,
|
| 849 |
+
"learning_rate": 9.519168849742603e-06,
|
| 850 |
+
"loss": 0.6536,
|
| 851 |
+
"step": 170
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 30.90909090909091,
|
| 855 |
+
"eval_loss": 0.71826171875,
|
| 856 |
+
"eval_runtime": 24.3503,
|
| 857 |
+
"eval_samples_per_second": 0.986,
|
| 858 |
+
"eval_steps_per_second": 0.986,
|
| 859 |
+
"step": 170
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 31.272727272727273,
|
| 863 |
+
"grad_norm": 0.2845761775970459,
|
| 864 |
+
"learning_rate": 9.499058738432492e-06,
|
| 865 |
+
"loss": 0.6384,
|
| 866 |
+
"step": 172
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 31.636363636363637,
|
| 870 |
+
"grad_norm": 0.30121341347694397,
|
| 871 |
+
"learning_rate": 9.478558801197065e-06,
|
| 872 |
+
"loss": 0.6191,
|
| 873 |
+
"step": 174
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 32.0,
|
| 877 |
+
"grad_norm": 0.2854570746421814,
|
| 878 |
+
"learning_rate": 9.457670814276083e-06,
|
| 879 |
+
"loss": 0.6092,
|
| 880 |
+
"step": 176
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 32.0,
|
| 884 |
+
"eval_loss": 0.7163482308387756,
|
| 885 |
+
"eval_runtime": 24.3583,
|
| 886 |
+
"eval_samples_per_second": 0.985,
|
| 887 |
+
"eval_steps_per_second": 0.985,
|
| 888 |
+
"step": 176
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 32.36363636363637,
|
| 892 |
+
"grad_norm": 0.3030169904232025,
|
| 893 |
+
"learning_rate": 9.436396587532297e-06,
|
| 894 |
+
"loss": 0.5929,
|
| 895 |
+
"step": 178
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 32.72727272727273,
|
| 899 |
+
"grad_norm": 0.2960076928138733,
|
| 900 |
+
"learning_rate": 9.414737964294636e-06,
|
| 901 |
+
"loss": 0.621,
|
| 902 |
+
"step": 180
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 32.90909090909091,
|
| 906 |
+
"eval_loss": 0.7148606181144714,
|
| 907 |
+
"eval_runtime": 24.3579,
|
| 908 |
+
"eval_samples_per_second": 0.985,
|
| 909 |
+
"eval_steps_per_second": 0.985,
|
| 910 |
+
"step": 181
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 33.09090909090909,
|
| 914 |
+
"grad_norm": 0.3322046995162964,
|
| 915 |
+
"learning_rate": 9.392696821198488e-06,
|
| 916 |
+
"loss": 0.6048,
|
| 917 |
+
"step": 182
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 33.45454545454545,
|
| 921 |
+
"grad_norm": 0.335716187953949,
|
| 922 |
+
"learning_rate": 9.370275068023097e-06,
|
| 923 |
+
"loss": 0.6233,
|
| 924 |
+
"step": 184
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 33.81818181818182,
|
| 928 |
+
"grad_norm": 0.3075459599494934,
|
| 929 |
+
"learning_rate": 9.347474647526095e-06,
|
| 930 |
+
"loss": 0.5823,
|
| 931 |
+
"step": 186
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 34.0,
|
| 935 |
+
"eval_loss": 0.7143663763999939,
|
| 936 |
+
"eval_runtime": 24.3571,
|
| 937 |
+
"eval_samples_per_second": 0.985,
|
| 938 |
+
"eval_steps_per_second": 0.985,
|
| 939 |
+
"step": 187
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 34.18181818181818,
|
| 943 |
+
"grad_norm": 0.3422604203224182,
|
| 944 |
+
"learning_rate": 9.324297535275156e-06,
|
| 945 |
+
"loss": 0.5754,
|
| 946 |
+
"step": 188
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 34.54545454545455,
|
| 950 |
+
"grad_norm": 0.33866289258003235,
|
| 951 |
+
"learning_rate": 9.30074573947683e-06,
|
| 952 |
+
"loss": 0.5951,
|
| 953 |
+
"step": 190
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 34.90909090909091,
|
| 957 |
+
"grad_norm": 0.37743914127349854,
|
| 958 |
+
"learning_rate": 9.276821300802535e-06,
|
| 959 |
+
"loss": 0.5651,
|
| 960 |
+
"step": 192
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 34.90909090909091,
|
| 964 |
+
"eval_loss": 0.7156122326850891,
|
| 965 |
+
"eval_runtime": 24.3497,
|
| 966 |
+
"eval_samples_per_second": 0.986,
|
| 967 |
+
"eval_steps_per_second": 0.986,
|
| 968 |
+
"step": 192
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 35.27272727272727,
|
| 972 |
+
"grad_norm": 0.3403712511062622,
|
| 973 |
+
"learning_rate": 9.25252629221175e-06,
|
| 974 |
+
"loss": 0.5877,
|
| 975 |
+
"step": 194
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 35.63636363636363,
|
| 979 |
+
"grad_norm": 0.32227441668510437,
|
| 980 |
+
"learning_rate": 9.227862818772392e-06,
|
| 981 |
+
"loss": 0.5569,
|
| 982 |
+
"step": 196
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 36.0,
|
| 986 |
+
"grad_norm": 0.3595390021800995,
|
| 987 |
+
"learning_rate": 9.202833017478421e-06,
|
| 988 |
+
"loss": 0.5951,
|
| 989 |
+
"step": 198
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 36.0,
|
| 993 |
+
"eval_loss": 0.7163762450218201,
|
| 994 |
+
"eval_runtime": 24.3582,
|
| 995 |
+
"eval_samples_per_second": 0.985,
|
| 996 |
+
"eval_steps_per_second": 0.985,
|
| 997 |
+
"step": 198
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 36.36363636363637,
|
| 1001 |
+
"grad_norm": 0.3400909900665283,
|
| 1002 |
+
"learning_rate": 9.177439057064684e-06,
|
| 1003 |
+
"loss": 0.552,
|
| 1004 |
+
"step": 200
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 36.72727272727273,
|
| 1008 |
+
"grad_norm": 0.3983656167984009,
|
| 1009 |
+
"learning_rate": 9.151683137818989e-06,
|
| 1010 |
+
"loss": 0.5637,
|
| 1011 |
+
"step": 202
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 36.90909090909091,
|
| 1015 |
+
"eval_loss": 0.7194623947143555,
|
| 1016 |
+
"eval_runtime": 24.3553,
|
| 1017 |
+
"eval_samples_per_second": 0.985,
|
| 1018 |
+
"eval_steps_per_second": 0.985,
|
| 1019 |
+
"step": 203
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 37.09090909090909,
|
| 1023 |
+
"grad_norm": 0.4176684319972992,
|
| 1024 |
+
"learning_rate": 9.125567491391476e-06,
|
| 1025 |
+
"loss": 0.5684,
|
| 1026 |
+
"step": 204
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 37.45454545454545,
|
| 1030 |
+
"grad_norm": 0.3539072275161743,
|
| 1031 |
+
"learning_rate": 9.099094380601244e-06,
|
| 1032 |
+
"loss": 0.5392,
|
| 1033 |
+
"step": 206
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 37.81818181818182,
|
| 1037 |
+
"grad_norm": 0.40454044938087463,
|
| 1038 |
+
"learning_rate": 9.072266099240286e-06,
|
| 1039 |
+
"loss": 0.5669,
|
| 1040 |
+
"step": 208
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 38.0,
|
| 1044 |
+
"eval_loss": 0.7218795418739319,
|
| 1045 |
+
"eval_runtime": 24.3543,
|
| 1046 |
+
"eval_samples_per_second": 0.985,
|
| 1047 |
+
"eval_steps_per_second": 0.985,
|
| 1048 |
+
"step": 209
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 38.18181818181818,
|
| 1052 |
+
"grad_norm": 0.3737054765224457,
|
| 1053 |
+
"learning_rate": 9.045084971874738e-06,
|
| 1054 |
+
"loss": 0.5348,
|
| 1055 |
+
"step": 210
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 38.54545454545455,
|
| 1059 |
+
"grad_norm": 0.37676355242729187,
|
| 1060 |
+
"learning_rate": 9.017553353643479e-06,
|
| 1061 |
+
"loss": 0.5322,
|
| 1062 |
+
"step": 212
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 38.90909090909091,
|
| 1066 |
+
"grad_norm": 0.41781264543533325,
|
| 1067 |
+
"learning_rate": 8.989673630054044e-06,
|
| 1068 |
+
"loss": 0.5613,
|
| 1069 |
+
"step": 214
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 38.90909090909091,
|
| 1073 |
+
"eval_loss": 0.7278082966804504,
|
| 1074 |
+
"eval_runtime": 24.3537,
|
| 1075 |
+
"eval_samples_per_second": 0.985,
|
| 1076 |
+
"eval_steps_per_second": 0.985,
|
| 1077 |
+
"step": 214
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 39.27272727272727,
|
| 1081 |
+
"grad_norm": 0.41486313939094543,
|
| 1082 |
+
"learning_rate": 8.961448216775955e-06,
|
| 1083 |
+
"loss": 0.5221,
|
| 1084 |
+
"step": 216
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 39.63636363636363,
|
| 1088 |
+
"grad_norm": 0.40721336007118225,
|
| 1089 |
+
"learning_rate": 8.932879559431392e-06,
|
| 1090 |
+
"loss": 0.5181,
|
| 1091 |
+
"step": 218
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 40.0,
|
| 1095 |
+
"grad_norm": 0.46421483159065247,
|
| 1096 |
+
"learning_rate": 8.903970133383297e-06,
|
| 1097 |
+
"loss": 0.5156,
|
| 1098 |
+
"step": 220
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 40.0,
|
| 1102 |
+
"eval_loss": 0.7309436798095703,
|
| 1103 |
+
"eval_runtime": 24.3653,
|
| 1104 |
+
"eval_samples_per_second": 0.985,
|
| 1105 |
+
"eval_steps_per_second": 0.985,
|
| 1106 |
+
"step": 220
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 40.36363636363637,
|
| 1110 |
+
"grad_norm": 0.5447797179222107,
|
| 1111 |
+
"learning_rate": 8.874722443520898e-06,
|
| 1112 |
+
"loss": 0.5196,
|
| 1113 |
+
"step": 222
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 40.72727272727273,
|
| 1117 |
+
"grad_norm": 0.44560515880584717,
|
| 1118 |
+
"learning_rate": 8.845139024042664e-06,
|
| 1119 |
+
"loss": 0.5044,
|
| 1120 |
+
"step": 224
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 40.90909090909091,
|
| 1124 |
+
"eval_loss": 0.739515483379364,
|
| 1125 |
+
"eval_runtime": 24.3585,
|
| 1126 |
+
"eval_samples_per_second": 0.985,
|
| 1127 |
+
"eval_steps_per_second": 0.985,
|
| 1128 |
+
"step": 225
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 40.90909090909091,
|
| 1132 |
+
"step": 225,
|
| 1133 |
+
"total_flos": 3.429455470966866e+17,
|
| 1134 |
+
"train_loss": 0.7736989471647474,
|
| 1135 |
+
"train_runtime": 12307.7975,
|
| 1136 |
+
"train_samples_per_second": 1.072,
|
| 1137 |
+
"train_steps_per_second": 0.061
|
| 1138 |
+
}
|
| 1139 |
+
],
|
| 1140 |
+
"logging_steps": 2,
|
| 1141 |
+
"max_steps": 750,
|
| 1142 |
+
"num_input_tokens_seen": 0,
|
| 1143 |
+
"num_train_epochs": 150,
|
| 1144 |
+
"save_steps": 25,
|
| 1145 |
+
"stateful_callbacks": {
|
| 1146 |
+
"EarlyStoppingCallback": {
|
| 1147 |
+
"args": {
|
| 1148 |
+
"early_stopping_patience": 7,
|
| 1149 |
+
"early_stopping_threshold": 0.0
|
| 1150 |
+
},
|
| 1151 |
+
"attributes": {
|
| 1152 |
+
"early_stopping_patience_counter": 0
|
| 1153 |
+
}
|
| 1154 |
+
},
|
| 1155 |
+
"TrainerControl": {
|
| 1156 |
+
"args": {
|
| 1157 |
+
"should_epoch_stop": false,
|
| 1158 |
+
"should_evaluate": false,
|
| 1159 |
+
"should_log": false,
|
| 1160 |
+
"should_save": true,
|
| 1161 |
+
"should_training_stop": true
|
| 1162 |
+
},
|
| 1163 |
+
"attributes": {}
|
| 1164 |
+
}
|
| 1165 |
+
},
|
| 1166 |
+
"total_flos": 3.429455470966866e+17,
|
| 1167 |
+
"train_batch_size": 1,
|
| 1168 |
+
"trial_name": null,
|
| 1169 |
+
"trial_params": null
|
| 1170 |
+
}
|