lillian039
commited on
Commit
•
66dba0f
1
Parent(s):
45a679b
Model save
Browse files- README.md +69 -0
- all_results.json +9 -0
- train_results.json +9 -0
- trainer_state.json +3111 -0
README.md
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
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: induction-20k-100seeds-barc-llama3.1-8b-instruct-lora64-induction-gpt4omini_lr2e-4_epoch3
|
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 |
+
# induction-20k-100seeds-barc-llama3.1-8b-instruct-lora64-induction-gpt4omini_lr2e-4_epoch3
|
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 an unknown dataset.
|
20 |
+
It achieves the following results on the evaluation set:
|
21 |
+
- Loss: 0.3362
|
22 |
+
|
23 |
+
## Model description
|
24 |
+
|
25 |
+
More information needed
|
26 |
+
|
27 |
+
## Intended uses & limitations
|
28 |
+
|
29 |
+
More information needed
|
30 |
+
|
31 |
+
## Training and evaluation data
|
32 |
+
|
33 |
+
More information needed
|
34 |
+
|
35 |
+
## Training procedure
|
36 |
+
|
37 |
+
### Training hyperparameters
|
38 |
+
|
39 |
+
The following hyperparameters were used during training:
|
40 |
+
- learning_rate: 0.0002
|
41 |
+
- train_batch_size: 8
|
42 |
+
- eval_batch_size: 4
|
43 |
+
- seed: 42
|
44 |
+
- distributed_type: multi-GPU
|
45 |
+
- num_devices: 8
|
46 |
+
- gradient_accumulation_steps: 2
|
47 |
+
- total_train_batch_size: 128
|
48 |
+
- total_eval_batch_size: 32
|
49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
50 |
+
- lr_scheduler_type: cosine
|
51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
52 |
+
- num_epochs: 3
|
53 |
+
|
54 |
+
### Training results
|
55 |
+
|
56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
57 |
+
|:-------------:|:------:|:----:|:---------------:|
|
58 |
+
| 0.3614 | 0.9966 | 145 | 0.3568 |
|
59 |
+
| 0.3258 | 2.0 | 291 | 0.3392 |
|
60 |
+
| 0.3019 | 2.9897 | 435 | 0.3362 |
|
61 |
+
|
62 |
+
|
63 |
+
### Framework versions
|
64 |
+
|
65 |
+
- PEFT 0.13.0
|
66 |
+
- Transformers 4.45.0.dev0
|
67 |
+
- Pytorch 2.4.0+cu121
|
68 |
+
- Datasets 3.0.1
|
69 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.9896907216494846,
|
3 |
+
"total_flos": 1.566236469816197e+17,
|
4 |
+
"train_loss": 0.3695115421695271,
|
5 |
+
"train_runtime": 4901.7941,
|
6 |
+
"train_samples": 18598,
|
7 |
+
"train_samples_per_second": 11.382,
|
8 |
+
"train_steps_per_second": 0.089
|
9 |
+
}
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 2.9896907216494846,
|
3 |
+
"total_flos": 1.566236469816197e+17,
|
4 |
+
"train_loss": 0.3695115421695271,
|
5 |
+
"train_runtime": 4901.7941,
|
6 |
+
"train_samples": 18598,
|
7 |
+
"train_samples_per_second": 11.382,
|
8 |
+
"train_steps_per_second": 0.089
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,3111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.9896907216494846,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 435,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.006872852233676976,
|
13 |
+
"grad_norm": 0.9025488538333486,
|
14 |
+
"learning_rate": 4.5454545454545455e-06,
|
15 |
+
"loss": 1.3348,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.013745704467353952,
|
20 |
+
"grad_norm": 0.90579592679034,
|
21 |
+
"learning_rate": 9.090909090909091e-06,
|
22 |
+
"loss": 1.3289,
|
23 |
+
"step": 2
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.020618556701030927,
|
27 |
+
"grad_norm": 0.8989699710739039,
|
28 |
+
"learning_rate": 1.3636363636363637e-05,
|
29 |
+
"loss": 1.3281,
|
30 |
+
"step": 3
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.027491408934707903,
|
34 |
+
"grad_norm": 0.8870740843172286,
|
35 |
+
"learning_rate": 1.8181818181818182e-05,
|
36 |
+
"loss": 1.3243,
|
37 |
+
"step": 4
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.03436426116838488,
|
41 |
+
"grad_norm": 0.8649293036935595,
|
42 |
+
"learning_rate": 2.272727272727273e-05,
|
43 |
+
"loss": 1.2973,
|
44 |
+
"step": 5
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.041237113402061855,
|
48 |
+
"grad_norm": 0.8290702683406395,
|
49 |
+
"learning_rate": 2.7272727272727273e-05,
|
50 |
+
"loss": 1.2932,
|
51 |
+
"step": 6
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.048109965635738834,
|
55 |
+
"grad_norm": 0.7168394097732069,
|
56 |
+
"learning_rate": 3.181818181818182e-05,
|
57 |
+
"loss": 1.2007,
|
58 |
+
"step": 7
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.054982817869415807,
|
62 |
+
"grad_norm": 0.5725102411275758,
|
63 |
+
"learning_rate": 3.6363636363636364e-05,
|
64 |
+
"loss": 1.1386,
|
65 |
+
"step": 8
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.061855670103092786,
|
69 |
+
"grad_norm": 0.496838314317771,
|
70 |
+
"learning_rate": 4.0909090909090915e-05,
|
71 |
+
"loss": 1.0441,
|
72 |
+
"step": 9
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.06872852233676977,
|
76 |
+
"grad_norm": 0.5142178206237444,
|
77 |
+
"learning_rate": 4.545454545454546e-05,
|
78 |
+
"loss": 1.0122,
|
79 |
+
"step": 10
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.07560137457044673,
|
83 |
+
"grad_norm": 0.5848893147382296,
|
84 |
+
"learning_rate": 5e-05,
|
85 |
+
"loss": 0.9549,
|
86 |
+
"step": 11
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.08247422680412371,
|
90 |
+
"grad_norm": 0.5616535995975048,
|
91 |
+
"learning_rate": 5.4545454545454546e-05,
|
92 |
+
"loss": 0.8907,
|
93 |
+
"step": 12
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.08934707903780069,
|
97 |
+
"grad_norm": 0.5769972107216242,
|
98 |
+
"learning_rate": 5.90909090909091e-05,
|
99 |
+
"loss": 0.8259,
|
100 |
+
"step": 13
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.09621993127147767,
|
104 |
+
"grad_norm": 0.5119411605843678,
|
105 |
+
"learning_rate": 6.363636363636364e-05,
|
106 |
+
"loss": 0.7488,
|
107 |
+
"step": 14
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.10309278350515463,
|
111 |
+
"grad_norm": 0.5355943194412696,
|
112 |
+
"learning_rate": 6.818181818181818e-05,
|
113 |
+
"loss": 0.7214,
|
114 |
+
"step": 15
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.10996563573883161,
|
118 |
+
"grad_norm": 0.43451997380890967,
|
119 |
+
"learning_rate": 7.272727272727273e-05,
|
120 |
+
"loss": 0.6451,
|
121 |
+
"step": 16
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.11683848797250859,
|
125 |
+
"grad_norm": 0.35905987630855185,
|
126 |
+
"learning_rate": 7.727272727272727e-05,
|
127 |
+
"loss": 0.6062,
|
128 |
+
"step": 17
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.12371134020618557,
|
132 |
+
"grad_norm": 0.2721170077931892,
|
133 |
+
"learning_rate": 8.181818181818183e-05,
|
134 |
+
"loss": 0.5685,
|
135 |
+
"step": 18
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.13058419243986255,
|
139 |
+
"grad_norm": 0.26373379486773657,
|
140 |
+
"learning_rate": 8.636363636363637e-05,
|
141 |
+
"loss": 0.5502,
|
142 |
+
"step": 19
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.13745704467353953,
|
146 |
+
"grad_norm": 0.22302209083632957,
|
147 |
+
"learning_rate": 9.090909090909092e-05,
|
148 |
+
"loss": 0.5313,
|
149 |
+
"step": 20
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.14432989690721648,
|
153 |
+
"grad_norm": 0.25718099226437285,
|
154 |
+
"learning_rate": 9.545454545454546e-05,
|
155 |
+
"loss": 0.5112,
|
156 |
+
"step": 21
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.15120274914089346,
|
160 |
+
"grad_norm": 0.2099769070901816,
|
161 |
+
"learning_rate": 0.0001,
|
162 |
+
"loss": 0.5062,
|
163 |
+
"step": 22
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.15807560137457044,
|
167 |
+
"grad_norm": 0.22657475295381196,
|
168 |
+
"learning_rate": 0.00010454545454545455,
|
169 |
+
"loss": 0.4993,
|
170 |
+
"step": 23
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.16494845360824742,
|
174 |
+
"grad_norm": 0.20635510069704832,
|
175 |
+
"learning_rate": 0.00010909090909090909,
|
176 |
+
"loss": 0.5045,
|
177 |
+
"step": 24
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.1718213058419244,
|
181 |
+
"grad_norm": 0.18926897428760356,
|
182 |
+
"learning_rate": 0.00011363636363636365,
|
183 |
+
"loss": 0.5024,
|
184 |
+
"step": 25
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.17869415807560138,
|
188 |
+
"grad_norm": 0.14996792404825188,
|
189 |
+
"learning_rate": 0.0001181818181818182,
|
190 |
+
"loss": 0.4682,
|
191 |
+
"step": 26
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.18556701030927836,
|
195 |
+
"grad_norm": 0.1496872862950363,
|
196 |
+
"learning_rate": 0.00012272727272727272,
|
197 |
+
"loss": 0.4709,
|
198 |
+
"step": 27
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.19243986254295534,
|
202 |
+
"grad_norm": 0.13209046184890866,
|
203 |
+
"learning_rate": 0.00012727272727272728,
|
204 |
+
"loss": 0.47,
|
205 |
+
"step": 28
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.19931271477663232,
|
209 |
+
"grad_norm": 0.12109603803428741,
|
210 |
+
"learning_rate": 0.0001318181818181818,
|
211 |
+
"loss": 0.4514,
|
212 |
+
"step": 29
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.20618556701030927,
|
216 |
+
"grad_norm": 0.11809201542659267,
|
217 |
+
"learning_rate": 0.00013636363636363637,
|
218 |
+
"loss": 0.4511,
|
219 |
+
"step": 30
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.21305841924398625,
|
223 |
+
"grad_norm": 0.11141685622199149,
|
224 |
+
"learning_rate": 0.00014090909090909093,
|
225 |
+
"loss": 0.4563,
|
226 |
+
"step": 31
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.21993127147766323,
|
230 |
+
"grad_norm": 0.11949577115206655,
|
231 |
+
"learning_rate": 0.00014545454545454546,
|
232 |
+
"loss": 0.4492,
|
233 |
+
"step": 32
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.2268041237113402,
|
237 |
+
"grad_norm": 0.11408053439939173,
|
238 |
+
"learning_rate": 0.00015000000000000001,
|
239 |
+
"loss": 0.4633,
|
240 |
+
"step": 33
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.23367697594501718,
|
244 |
+
"grad_norm": 0.1062170565004825,
|
245 |
+
"learning_rate": 0.00015454545454545454,
|
246 |
+
"loss": 0.4537,
|
247 |
+
"step": 34
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.24054982817869416,
|
251 |
+
"grad_norm": 0.1031396270785207,
|
252 |
+
"learning_rate": 0.0001590909090909091,
|
253 |
+
"loss": 0.4293,
|
254 |
+
"step": 35
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.24742268041237114,
|
258 |
+
"grad_norm": 0.1138852883627702,
|
259 |
+
"learning_rate": 0.00016363636363636366,
|
260 |
+
"loss": 0.4362,
|
261 |
+
"step": 36
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.2542955326460481,
|
265 |
+
"grad_norm": 0.10644832100598535,
|
266 |
+
"learning_rate": 0.0001681818181818182,
|
267 |
+
"loss": 0.4415,
|
268 |
+
"step": 37
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.2611683848797251,
|
272 |
+
"grad_norm": 0.10345533799672353,
|
273 |
+
"learning_rate": 0.00017272727272727275,
|
274 |
+
"loss": 0.436,
|
275 |
+
"step": 38
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.26804123711340205,
|
279 |
+
"grad_norm": 0.10610064113960643,
|
280 |
+
"learning_rate": 0.00017727272727272728,
|
281 |
+
"loss": 0.4287,
|
282 |
+
"step": 39
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.27491408934707906,
|
286 |
+
"grad_norm": 0.10969105942235169,
|
287 |
+
"learning_rate": 0.00018181818181818183,
|
288 |
+
"loss": 0.4338,
|
289 |
+
"step": 40
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.281786941580756,
|
293 |
+
"grad_norm": 0.0970601684543591,
|
294 |
+
"learning_rate": 0.00018636363636363636,
|
295 |
+
"loss": 0.4006,
|
296 |
+
"step": 41
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.28865979381443296,
|
300 |
+
"grad_norm": 0.10381671981330426,
|
301 |
+
"learning_rate": 0.00019090909090909092,
|
302 |
+
"loss": 0.4273,
|
303 |
+
"step": 42
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.29553264604810997,
|
307 |
+
"grad_norm": 0.09976207846850071,
|
308 |
+
"learning_rate": 0.00019545454545454548,
|
309 |
+
"loss": 0.4235,
|
310 |
+
"step": 43
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.3024054982817869,
|
314 |
+
"grad_norm": 0.09816614528049827,
|
315 |
+
"learning_rate": 0.0002,
|
316 |
+
"loss": 0.4214,
|
317 |
+
"step": 44
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.30927835051546393,
|
321 |
+
"grad_norm": 0.09512891102342932,
|
322 |
+
"learning_rate": 0.00019999677214588312,
|
323 |
+
"loss": 0.4123,
|
324 |
+
"step": 45
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.3161512027491409,
|
328 |
+
"grad_norm": 0.09712188213416782,
|
329 |
+
"learning_rate": 0.00019998708879191335,
|
330 |
+
"loss": 0.4113,
|
331 |
+
"step": 46
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.3230240549828179,
|
335 |
+
"grad_norm": 0.09165233501961581,
|
336 |
+
"learning_rate": 0.00019997095056321971,
|
337 |
+
"loss": 0.4058,
|
338 |
+
"step": 47
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.32989690721649484,
|
342 |
+
"grad_norm": 0.09882634020532893,
|
343 |
+
"learning_rate": 0.00019994835850163924,
|
344 |
+
"loss": 0.4178,
|
345 |
+
"step": 48
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.33676975945017185,
|
349 |
+
"grad_norm": 0.09028464086542946,
|
350 |
+
"learning_rate": 0.00019991931406564944,
|
351 |
+
"loss": 0.4089,
|
352 |
+
"step": 49
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.3436426116838488,
|
356 |
+
"grad_norm": 0.09886226546141882,
|
357 |
+
"learning_rate": 0.00019988381913027442,
|
358 |
+
"loss": 0.4078,
|
359 |
+
"step": 50
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.35051546391752575,
|
363 |
+
"grad_norm": 0.10025332685705594,
|
364 |
+
"learning_rate": 0.00019984187598696363,
|
365 |
+
"loss": 0.3986,
|
366 |
+
"step": 51
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.35738831615120276,
|
370 |
+
"grad_norm": 0.08707000383596354,
|
371 |
+
"learning_rate": 0.00019979348734344398,
|
372 |
+
"loss": 0.3872,
|
373 |
+
"step": 52
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.3642611683848797,
|
377 |
+
"grad_norm": 0.08932651653882982,
|
378 |
+
"learning_rate": 0.00019973865632354516,
|
379 |
+
"loss": 0.4015,
|
380 |
+
"step": 53
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.3711340206185567,
|
384 |
+
"grad_norm": 0.08566831385278742,
|
385 |
+
"learning_rate": 0.0001996773864669978,
|
386 |
+
"loss": 0.3992,
|
387 |
+
"step": 54
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.37800687285223367,
|
391 |
+
"grad_norm": 0.09225751803383826,
|
392 |
+
"learning_rate": 0.00019960968172920516,
|
393 |
+
"loss": 0.4032,
|
394 |
+
"step": 55
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.3848797250859107,
|
398 |
+
"grad_norm": 0.08842990105324244,
|
399 |
+
"learning_rate": 0.00019953554648098748,
|
400 |
+
"loss": 0.3847,
|
401 |
+
"step": 56
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.3917525773195876,
|
405 |
+
"grad_norm": 0.08870257839882814,
|
406 |
+
"learning_rate": 0.0001994549855083001,
|
407 |
+
"loss": 0.397,
|
408 |
+
"step": 57
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.39862542955326463,
|
412 |
+
"grad_norm": 0.08786210264099557,
|
413 |
+
"learning_rate": 0.0001993680040119244,
|
414 |
+
"loss": 0.3974,
|
415 |
+
"step": 58
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.4054982817869416,
|
419 |
+
"grad_norm": 0.0875901217734107,
|
420 |
+
"learning_rate": 0.00019927460760713197,
|
421 |
+
"loss": 0.3828,
|
422 |
+
"step": 59
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.41237113402061853,
|
426 |
+
"grad_norm": 0.0915443237168321,
|
427 |
+
"learning_rate": 0.00019917480232332224,
|
428 |
+
"loss": 0.3874,
|
429 |
+
"step": 60
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.41924398625429554,
|
433 |
+
"grad_norm": 0.08637492208504707,
|
434 |
+
"learning_rate": 0.00019906859460363307,
|
435 |
+
"loss": 0.3842,
|
436 |
+
"step": 61
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.4261168384879725,
|
440 |
+
"grad_norm": 0.08978496785747528,
|
441 |
+
"learning_rate": 0.00019895599130452505,
|
442 |
+
"loss": 0.3808,
|
443 |
+
"step": 62
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.4329896907216495,
|
447 |
+
"grad_norm": 0.09345275391530274,
|
448 |
+
"learning_rate": 0.0001988369996953386,
|
449 |
+
"loss": 0.3896,
|
450 |
+
"step": 63
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.43986254295532645,
|
454 |
+
"grad_norm": 0.08240749135703576,
|
455 |
+
"learning_rate": 0.00019871162745782478,
|
456 |
+
"loss": 0.3846,
|
457 |
+
"step": 64
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.44673539518900346,
|
461 |
+
"grad_norm": 0.08438226126381838,
|
462 |
+
"learning_rate": 0.00019857988268564953,
|
463 |
+
"loss": 0.3887,
|
464 |
+
"step": 65
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.4536082474226804,
|
468 |
+
"grad_norm": 0.09398865182960808,
|
469 |
+
"learning_rate": 0.0001984417738838709,
|
470 |
+
"loss": 0.3818,
|
471 |
+
"step": 66
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.46048109965635736,
|
475 |
+
"grad_norm": 0.0908839241950218,
|
476 |
+
"learning_rate": 0.0001982973099683902,
|
477 |
+
"loss": 0.3773,
|
478 |
+
"step": 67
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.46735395189003437,
|
482 |
+
"grad_norm": 0.09531921092274791,
|
483 |
+
"learning_rate": 0.0001981465002653763,
|
484 |
+
"loss": 0.3847,
|
485 |
+
"step": 68
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.4742268041237113,
|
489 |
+
"grad_norm": 0.09706258996994553,
|
490 |
+
"learning_rate": 0.00019798935451066361,
|
491 |
+
"loss": 0.3907,
|
492 |
+
"step": 69
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.48109965635738833,
|
496 |
+
"grad_norm": 0.09575939448299316,
|
497 |
+
"learning_rate": 0.0001978258828491236,
|
498 |
+
"loss": 0.3876,
|
499 |
+
"step": 70
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.4879725085910653,
|
503 |
+
"grad_norm": 0.09521084352265623,
|
504 |
+
"learning_rate": 0.00019765609583400977,
|
505 |
+
"loss": 0.3892,
|
506 |
+
"step": 71
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.4948453608247423,
|
510 |
+
"grad_norm": 0.09209593500458252,
|
511 |
+
"learning_rate": 0.0001974800044262764,
|
512 |
+
"loss": 0.3797,
|
513 |
+
"step": 72
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.5017182130584192,
|
517 |
+
"grad_norm": 0.09988994775547788,
|
518 |
+
"learning_rate": 0.00019729761999387103,
|
519 |
+
"loss": 0.3799,
|
520 |
+
"step": 73
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.5085910652920962,
|
524 |
+
"grad_norm": 0.0925159404721338,
|
525 |
+
"learning_rate": 0.00019710895431100046,
|
526 |
+
"loss": 0.3709,
|
527 |
+
"step": 74
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.5154639175257731,
|
531 |
+
"grad_norm": 0.08980873486318591,
|
532 |
+
"learning_rate": 0.00019691401955737072,
|
533 |
+
"loss": 0.3701,
|
534 |
+
"step": 75
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.5223367697594502,
|
538 |
+
"grad_norm": 0.09839291302100885,
|
539 |
+
"learning_rate": 0.00019671282831740076,
|
540 |
+
"loss": 0.377,
|
541 |
+
"step": 76
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.5292096219931272,
|
545 |
+
"grad_norm": 0.10101389839072905,
|
546 |
+
"learning_rate": 0.00019650539357941003,
|
547 |
+
"loss": 0.3848,
|
548 |
+
"step": 77
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.5360824742268041,
|
552 |
+
"grad_norm": 0.09386410727988179,
|
553 |
+
"learning_rate": 0.00019629172873477995,
|
554 |
+
"loss": 0.3957,
|
555 |
+
"step": 78
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.5429553264604811,
|
559 |
+
"grad_norm": 0.09965731195409483,
|
560 |
+
"learning_rate": 0.00019607184757708951,
|
561 |
+
"loss": 0.3663,
|
562 |
+
"step": 79
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.5498281786941581,
|
566 |
+
"grad_norm": 0.09718899466523188,
|
567 |
+
"learning_rate": 0.00019584576430122473,
|
568 |
+
"loss": 0.3815,
|
569 |
+
"step": 80
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.5567010309278351,
|
573 |
+
"grad_norm": 0.08670051499783159,
|
574 |
+
"learning_rate": 0.00019561349350246226,
|
575 |
+
"loss": 0.3571,
|
576 |
+
"step": 81
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.563573883161512,
|
580 |
+
"grad_norm": 0.09482282486693373,
|
581 |
+
"learning_rate": 0.00019537505017552716,
|
582 |
+
"loss": 0.3755,
|
583 |
+
"step": 82
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.570446735395189,
|
587 |
+
"grad_norm": 0.09121809486362457,
|
588 |
+
"learning_rate": 0.00019513044971362494,
|
589 |
+
"loss": 0.3699,
|
590 |
+
"step": 83
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.5773195876288659,
|
594 |
+
"grad_norm": 0.084539930216277,
|
595 |
+
"learning_rate": 0.00019487970790744774,
|
596 |
+
"loss": 0.3638,
|
597 |
+
"step": 84
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.584192439862543,
|
601 |
+
"grad_norm": 0.09738298062170189,
|
602 |
+
"learning_rate": 0.000194622840944155,
|
603 |
+
"loss": 0.3727,
|
604 |
+
"step": 85
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.5910652920962199,
|
608 |
+
"grad_norm": 0.08993070175690983,
|
609 |
+
"learning_rate": 0.00019435986540632843,
|
610 |
+
"loss": 0.3579,
|
611 |
+
"step": 86
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.5979381443298969,
|
615 |
+
"grad_norm": 0.09867363864736761,
|
616 |
+
"learning_rate": 0.00019409079827090145,
|
617 |
+
"loss": 0.3733,
|
618 |
+
"step": 87
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.6048109965635738,
|
622 |
+
"grad_norm": 0.09589440600723866,
|
623 |
+
"learning_rate": 0.00019381565690806328,
|
624 |
+
"loss": 0.3763,
|
625 |
+
"step": 88
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.6116838487972509,
|
629 |
+
"grad_norm": 0.08659099561537051,
|
630 |
+
"learning_rate": 0.00019353445908013755,
|
631 |
+
"loss": 0.3639,
|
632 |
+
"step": 89
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.6185567010309279,
|
636 |
+
"grad_norm": 0.09401775227248314,
|
637 |
+
"learning_rate": 0.00019324722294043558,
|
638 |
+
"loss": 0.375,
|
639 |
+
"step": 90
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.6254295532646048,
|
643 |
+
"grad_norm": 0.08996648986031713,
|
644 |
+
"learning_rate": 0.00019295396703208453,
|
645 |
+
"loss": 0.3792,
|
646 |
+
"step": 91
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.6323024054982818,
|
650 |
+
"grad_norm": 0.08895665807543628,
|
651 |
+
"learning_rate": 0.00019265471028683014,
|
652 |
+
"loss": 0.3695,
|
653 |
+
"step": 92
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.6391752577319587,
|
657 |
+
"grad_norm": 0.09144896250727937,
|
658 |
+
"learning_rate": 0.00019234947202381486,
|
659 |
+
"loss": 0.3595,
|
660 |
+
"step": 93
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.6460481099656358,
|
664 |
+
"grad_norm": 0.08887150310263081,
|
665 |
+
"learning_rate": 0.00019203827194833026,
|
666 |
+
"loss": 0.3604,
|
667 |
+
"step": 94
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.6529209621993127,
|
671 |
+
"grad_norm": 0.09117379245722905,
|
672 |
+
"learning_rate": 0.00019172113015054532,
|
673 |
+
"loss": 0.371,
|
674 |
+
"step": 95
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.6597938144329897,
|
678 |
+
"grad_norm": 0.10193933851868603,
|
679 |
+
"learning_rate": 0.00019139806710420914,
|
680 |
+
"loss": 0.3788,
|
681 |
+
"step": 96
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.6666666666666666,
|
685 |
+
"grad_norm": 0.09395998486088643,
|
686 |
+
"learning_rate": 0.00019106910366532942,
|
687 |
+
"loss": 0.3696,
|
688 |
+
"step": 97
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.6735395189003437,
|
692 |
+
"grad_norm": 0.08737703293117173,
|
693 |
+
"learning_rate": 0.000190734261070826,
|
694 |
+
"loss": 0.3702,
|
695 |
+
"step": 98
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.6804123711340206,
|
699 |
+
"grad_norm": 0.08548511699995505,
|
700 |
+
"learning_rate": 0.00019039356093715975,
|
701 |
+
"loss": 0.3657,
|
702 |
+
"step": 99
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.6872852233676976,
|
706 |
+
"grad_norm": 0.09604747876952273,
|
707 |
+
"learning_rate": 0.00019004702525893732,
|
708 |
+
"loss": 0.3549,
|
709 |
+
"step": 100
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.6941580756013745,
|
713 |
+
"grad_norm": 0.08846405067678506,
|
714 |
+
"learning_rate": 0.000189694676407491,
|
715 |
+
"loss": 0.3749,
|
716 |
+
"step": 101
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.7010309278350515,
|
720 |
+
"grad_norm": 0.08626613784782615,
|
721 |
+
"learning_rate": 0.0001893365371294346,
|
722 |
+
"loss": 0.3582,
|
723 |
+
"step": 102
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.7079037800687286,
|
727 |
+
"grad_norm": 0.09126057669481359,
|
728 |
+
"learning_rate": 0.00018897263054519498,
|
729 |
+
"loss": 0.3534,
|
730 |
+
"step": 103
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.7147766323024055,
|
734 |
+
"grad_norm": 0.0900454809710255,
|
735 |
+
"learning_rate": 0.00018860298014751944,
|
736 |
+
"loss": 0.3696,
|
737 |
+
"step": 104
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.7216494845360825,
|
741 |
+
"grad_norm": 0.08917839487824428,
|
742 |
+
"learning_rate": 0.0001882276097999592,
|
743 |
+
"loss": 0.3646,
|
744 |
+
"step": 105
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.7285223367697594,
|
748 |
+
"grad_norm": 0.0882391823659059,
|
749 |
+
"learning_rate": 0.00018784654373532866,
|
750 |
+
"loss": 0.3559,
|
751 |
+
"step": 106
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.7353951890034365,
|
755 |
+
"grad_norm": 0.09228143955410971,
|
756 |
+
"learning_rate": 0.00018745980655414114,
|
757 |
+
"loss": 0.3664,
|
758 |
+
"step": 107
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.7422680412371134,
|
762 |
+
"grad_norm": 0.08980802312721321,
|
763 |
+
"learning_rate": 0.00018706742322302064,
|
764 |
+
"loss": 0.3555,
|
765 |
+
"step": 108
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.7491408934707904,
|
769 |
+
"grad_norm": 0.0949338333058196,
|
770 |
+
"learning_rate": 0.00018666941907309026,
|
771 |
+
"loss": 0.3646,
|
772 |
+
"step": 109
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.7560137457044673,
|
776 |
+
"grad_norm": 0.09145846210894344,
|
777 |
+
"learning_rate": 0.0001862658197983366,
|
778 |
+
"loss": 0.3633,
|
779 |
+
"step": 110
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.7628865979381443,
|
783 |
+
"grad_norm": 0.09186458995446992,
|
784 |
+
"learning_rate": 0.0001858566514539513,
|
785 |
+
"loss": 0.3643,
|
786 |
+
"step": 111
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.7697594501718213,
|
790 |
+
"grad_norm": 0.0972669632146363,
|
791 |
+
"learning_rate": 0.00018544194045464886,
|
792 |
+
"loss": 0.3655,
|
793 |
+
"step": 112
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.7766323024054983,
|
797 |
+
"grad_norm": 0.09359471372767092,
|
798 |
+
"learning_rate": 0.00018502171357296144,
|
799 |
+
"loss": 0.3603,
|
800 |
+
"step": 113
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.7835051546391752,
|
804 |
+
"grad_norm": 0.09214156653558547,
|
805 |
+
"learning_rate": 0.0001845959979375104,
|
806 |
+
"loss": 0.364,
|
807 |
+
"step": 114
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.7903780068728522,
|
811 |
+
"grad_norm": 0.08444087757336682,
|
812 |
+
"learning_rate": 0.00018416482103125506,
|
813 |
+
"loss": 0.34,
|
814 |
+
"step": 115
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.7972508591065293,
|
818 |
+
"grad_norm": 0.09276507804743452,
|
819 |
+
"learning_rate": 0.0001837282106897185,
|
820 |
+
"loss": 0.368,
|
821 |
+
"step": 116
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.8041237113402062,
|
825 |
+
"grad_norm": 0.09378471167753966,
|
826 |
+
"learning_rate": 0.00018328619509919044,
|
827 |
+
"loss": 0.3482,
|
828 |
+
"step": 117
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.8109965635738832,
|
832 |
+
"grad_norm": 0.09332313431503723,
|
833 |
+
"learning_rate": 0.0001828388027949078,
|
834 |
+
"loss": 0.3641,
|
835 |
+
"step": 118
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.8178694158075601,
|
839 |
+
"grad_norm": 0.09305498701606231,
|
840 |
+
"learning_rate": 0.00018238606265921238,
|
841 |
+
"loss": 0.3751,
|
842 |
+
"step": 119
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.8247422680412371,
|
846 |
+
"grad_norm": 0.09016788292835455,
|
847 |
+
"learning_rate": 0.00018192800391968642,
|
848 |
+
"loss": 0.3618,
|
849 |
+
"step": 120
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.8316151202749141,
|
853 |
+
"grad_norm": 0.08990882038453075,
|
854 |
+
"learning_rate": 0.00018146465614726567,
|
855 |
+
"loss": 0.3648,
|
856 |
+
"step": 121
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.8384879725085911,
|
860 |
+
"grad_norm": 0.08730362939825315,
|
861 |
+
"learning_rate": 0.00018099604925433043,
|
862 |
+
"loss": 0.3648,
|
863 |
+
"step": 122
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.845360824742268,
|
867 |
+
"grad_norm": 0.0855529056427899,
|
868 |
+
"learning_rate": 0.00018052221349277442,
|
869 |
+
"loss": 0.3555,
|
870 |
+
"step": 123
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.852233676975945,
|
874 |
+
"grad_norm": 0.08483030758119947,
|
875 |
+
"learning_rate": 0.00018004317945205197,
|
876 |
+
"loss": 0.3445,
|
877 |
+
"step": 124
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.8591065292096219,
|
881 |
+
"grad_norm": 0.09141374941069849,
|
882 |
+
"learning_rate": 0.0001795589780572031,
|
883 |
+
"loss": 0.3578,
|
884 |
+
"step": 125
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.865979381443299,
|
888 |
+
"grad_norm": 0.09172449205019759,
|
889 |
+
"learning_rate": 0.00017906964056685706,
|
890 |
+
"loss": 0.369,
|
891 |
+
"step": 126
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.872852233676976,
|
895 |
+
"grad_norm": 0.09194476004557194,
|
896 |
+
"learning_rate": 0.00017857519857121458,
|
897 |
+
"loss": 0.3536,
|
898 |
+
"step": 127
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.8797250859106529,
|
902 |
+
"grad_norm": 0.08791373718301158,
|
903 |
+
"learning_rate": 0.00017807568399000822,
|
904 |
+
"loss": 0.359,
|
905 |
+
"step": 128
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.8865979381443299,
|
909 |
+
"grad_norm": 0.08763620600672854,
|
910 |
+
"learning_rate": 0.000177571129070442,
|
911 |
+
"loss": 0.3679,
|
912 |
+
"step": 129
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.8934707903780069,
|
916 |
+
"grad_norm": 0.09331900791881745,
|
917 |
+
"learning_rate": 0.0001770615663851093,
|
918 |
+
"loss": 0.3575,
|
919 |
+
"step": 130
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.9003436426116839,
|
923 |
+
"grad_norm": 0.08946848422927521,
|
924 |
+
"learning_rate": 0.0001765470288298905,
|
925 |
+
"loss": 0.3612,
|
926 |
+
"step": 131
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.9072164948453608,
|
930 |
+
"grad_norm": 0.08790299300002972,
|
931 |
+
"learning_rate": 0.0001760275496218288,
|
932 |
+
"loss": 0.3585,
|
933 |
+
"step": 132
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.9140893470790378,
|
937 |
+
"grad_norm": 0.08910732597794836,
|
938 |
+
"learning_rate": 0.0001755031622969862,
|
939 |
+
"loss": 0.3433,
|
940 |
+
"step": 133
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.9209621993127147,
|
944 |
+
"grad_norm": 0.09095549976010009,
|
945 |
+
"learning_rate": 0.00017497390070827848,
|
946 |
+
"loss": 0.3553,
|
947 |
+
"step": 134
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.9278350515463918,
|
951 |
+
"grad_norm": 0.08642234772548092,
|
952 |
+
"learning_rate": 0.00017443979902328956,
|
953 |
+
"loss": 0.345,
|
954 |
+
"step": 135
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.9347079037800687,
|
958 |
+
"grad_norm": 0.08722908773912248,
|
959 |
+
"learning_rate": 0.00017390089172206592,
|
960 |
+
"loss": 0.3549,
|
961 |
+
"step": 136
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.9415807560137457,
|
965 |
+
"grad_norm": 0.08479526812891544,
|
966 |
+
"learning_rate": 0.00017335721359489057,
|
967 |
+
"loss": 0.3551,
|
968 |
+
"step": 137
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.9484536082474226,
|
972 |
+
"grad_norm": 0.08383082069809095,
|
973 |
+
"learning_rate": 0.00017280879974003707,
|
974 |
+
"loss": 0.3477,
|
975 |
+
"step": 138
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.9553264604810997,
|
979 |
+
"grad_norm": 0.08855742642369993,
|
980 |
+
"learning_rate": 0.0001722556855615039,
|
981 |
+
"loss": 0.3443,
|
982 |
+
"step": 139
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.9621993127147767,
|
986 |
+
"grad_norm": 0.09383265999192292,
|
987 |
+
"learning_rate": 0.00017169790676672858,
|
988 |
+
"loss": 0.3758,
|
989 |
+
"step": 140
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.9690721649484536,
|
993 |
+
"grad_norm": 0.08810377563181306,
|
994 |
+
"learning_rate": 0.0001711354993642827,
|
995 |
+
"loss": 0.3532,
|
996 |
+
"step": 141
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.9759450171821306,
|
1000 |
+
"grad_norm": 0.08975849483077629,
|
1001 |
+
"learning_rate": 0.0001705684996615472,
|
1002 |
+
"loss": 0.3643,
|
1003 |
+
"step": 142
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.9828178694158075,
|
1007 |
+
"grad_norm": 0.09369454621504099,
|
1008 |
+
"learning_rate": 0.0001699969442623686,
|
1009 |
+
"loss": 0.3478,
|
1010 |
+
"step": 143
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.9896907216494846,
|
1014 |
+
"grad_norm": 0.09057986523356745,
|
1015 |
+
"learning_rate": 0.00016942087006469592,
|
1016 |
+
"loss": 0.3511,
|
1017 |
+
"step": 144
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.9965635738831615,
|
1021 |
+
"grad_norm": 0.09045137945001339,
|
1022 |
+
"learning_rate": 0.00016884031425819853,
|
1023 |
+
"loss": 0.3614,
|
1024 |
+
"step": 145
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.9965635738831615,
|
1028 |
+
"eval_loss": 0.3568178713321686,
|
1029 |
+
"eval_runtime": 35.2647,
|
1030 |
+
"eval_samples_per_second": 27.761,
|
1031 |
+
"eval_steps_per_second": 0.879,
|
1032 |
+
"step": 145
|
1033 |
+
},
|
1034 |
+
{
|
1035 |
+
"epoch": 1.0034364261168385,
|
1036 |
+
"grad_norm": 0.08866714568686841,
|
1037 |
+
"learning_rate": 0.00016825531432186543,
|
1038 |
+
"loss": 0.3498,
|
1039 |
+
"step": 146
|
1040 |
+
},
|
1041 |
+
{
|
1042 |
+
"epoch": 1.0103092783505154,
|
1043 |
+
"grad_norm": 0.08358584630474639,
|
1044 |
+
"learning_rate": 0.00016766590802158566,
|
1045 |
+
"loss": 0.3362,
|
1046 |
+
"step": 147
|
1047 |
+
},
|
1048 |
+
{
|
1049 |
+
"epoch": 1.0171821305841924,
|
1050 |
+
"grad_norm": 0.09550594874077385,
|
1051 |
+
"learning_rate": 0.0001670721334077103,
|
1052 |
+
"loss": 0.3324,
|
1053 |
+
"step": 148
|
1054 |
+
},
|
1055 |
+
{
|
1056 |
+
"epoch": 1.0240549828178693,
|
1057 |
+
"grad_norm": 0.08478527036729729,
|
1058 |
+
"learning_rate": 0.00016647402881259598,
|
1059 |
+
"loss": 0.3341,
|
1060 |
+
"step": 149
|
1061 |
+
},
|
1062 |
+
{
|
1063 |
+
"epoch": 1.0309278350515463,
|
1064 |
+
"grad_norm": 0.08830292112242692,
|
1065 |
+
"learning_rate": 0.00016587163284813032,
|
1066 |
+
"loss": 0.3473,
|
1067 |
+
"step": 150
|
1068 |
+
},
|
1069 |
+
{
|
1070 |
+
"epoch": 1.0378006872852235,
|
1071 |
+
"grad_norm": 0.08704005978825016,
|
1072 |
+
"learning_rate": 0.00016526498440323914,
|
1073 |
+
"loss": 0.3407,
|
1074 |
+
"step": 151
|
1075 |
+
},
|
1076 |
+
{
|
1077 |
+
"epoch": 1.0446735395189004,
|
1078 |
+
"grad_norm": 0.08601362071028429,
|
1079 |
+
"learning_rate": 0.0001646541226413761,
|
1080 |
+
"loss": 0.3363,
|
1081 |
+
"step": 152
|
1082 |
+
},
|
1083 |
+
{
|
1084 |
+
"epoch": 1.0515463917525774,
|
1085 |
+
"grad_norm": 0.0904598808631882,
|
1086 |
+
"learning_rate": 0.00016403908699799425,
|
1087 |
+
"loss": 0.3374,
|
1088 |
+
"step": 153
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 1.0584192439862543,
|
1092 |
+
"grad_norm": 0.08475241611249208,
|
1093 |
+
"learning_rate": 0.00016341991717800023,
|
1094 |
+
"loss": 0.3336,
|
1095 |
+
"step": 154
|
1096 |
+
},
|
1097 |
+
{
|
1098 |
+
"epoch": 1.0652920962199313,
|
1099 |
+
"grad_norm": 0.08482091083905234,
|
1100 |
+
"learning_rate": 0.00016279665315319114,
|
1101 |
+
"loss": 0.323,
|
1102 |
+
"step": 155
|
1103 |
+
},
|
1104 |
+
{
|
1105 |
+
"epoch": 1.0721649484536082,
|
1106 |
+
"grad_norm": 0.08726380155985979,
|
1107 |
+
"learning_rate": 0.0001621693351596739,
|
1108 |
+
"loss": 0.3424,
|
1109 |
+
"step": 156
|
1110 |
+
},
|
1111 |
+
{
|
1112 |
+
"epoch": 1.0790378006872852,
|
1113 |
+
"grad_norm": 0.0853015885123658,
|
1114 |
+
"learning_rate": 0.00016153800369526788,
|
1115 |
+
"loss": 0.3423,
|
1116 |
+
"step": 157
|
1117 |
+
},
|
1118 |
+
{
|
1119 |
+
"epoch": 1.0859106529209621,
|
1120 |
+
"grad_norm": 0.08805974235730266,
|
1121 |
+
"learning_rate": 0.0001609026995168904,
|
1122 |
+
"loss": 0.3502,
|
1123 |
+
"step": 158
|
1124 |
+
},
|
1125 |
+
{
|
1126 |
+
"epoch": 1.0927835051546393,
|
1127 |
+
"grad_norm": 0.09726311836076129,
|
1128 |
+
"learning_rate": 0.00016026346363792567,
|
1129 |
+
"loss": 0.349,
|
1130 |
+
"step": 159
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 1.0996563573883162,
|
1134 |
+
"grad_norm": 0.08595219306786538,
|
1135 |
+
"learning_rate": 0.00015962033732557686,
|
1136 |
+
"loss": 0.3341,
|
1137 |
+
"step": 160
|
1138 |
+
},
|
1139 |
+
{
|
1140 |
+
"epoch": 1.1065292096219932,
|
1141 |
+
"grad_norm": 0.08741020566424072,
|
1142 |
+
"learning_rate": 0.00015897336209820239,
|
1143 |
+
"loss": 0.3302,
|
1144 |
+
"step": 161
|
1145 |
+
},
|
1146 |
+
{
|
1147 |
+
"epoch": 1.1134020618556701,
|
1148 |
+
"grad_norm": 0.09065581289645924,
|
1149 |
+
"learning_rate": 0.00015832257972263523,
|
1150 |
+
"loss": 0.3525,
|
1151 |
+
"step": 162
|
1152 |
+
},
|
1153 |
+
{
|
1154 |
+
"epoch": 1.120274914089347,
|
1155 |
+
"grad_norm": 0.08789954415852048,
|
1156 |
+
"learning_rate": 0.00015766803221148673,
|
1157 |
+
"loss": 0.3401,
|
1158 |
+
"step": 163
|
1159 |
+
},
|
1160 |
+
{
|
1161 |
+
"epoch": 1.127147766323024,
|
1162 |
+
"grad_norm": 0.09263710437745579,
|
1163 |
+
"learning_rate": 0.0001570097618204345,
|
1164 |
+
"loss": 0.3527,
|
1165 |
+
"step": 164
|
1166 |
+
},
|
1167 |
+
{
|
1168 |
+
"epoch": 1.134020618556701,
|
1169 |
+
"grad_norm": 0.08829924428217856,
|
1170 |
+
"learning_rate": 0.00015634781104549442,
|
1171 |
+
"loss": 0.3362,
|
1172 |
+
"step": 165
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 1.140893470790378,
|
1176 |
+
"grad_norm": 0.09275098741146737,
|
1177 |
+
"learning_rate": 0.00015568222262027717,
|
1178 |
+
"loss": 0.3309,
|
1179 |
+
"step": 166
|
1180 |
+
},
|
1181 |
+
{
|
1182 |
+
"epoch": 1.147766323024055,
|
1183 |
+
"grad_norm": 0.08949383804799335,
|
1184 |
+
"learning_rate": 0.00015501303951322943,
|
1185 |
+
"loss": 0.3388,
|
1186 |
+
"step": 167
|
1187 |
+
},
|
1188 |
+
{
|
1189 |
+
"epoch": 1.1546391752577319,
|
1190 |
+
"grad_norm": 0.0911492998049104,
|
1191 |
+
"learning_rate": 0.00015434030492486023,
|
1192 |
+
"loss": 0.3368,
|
1193 |
+
"step": 168
|
1194 |
+
},
|
1195 |
+
{
|
1196 |
+
"epoch": 1.161512027491409,
|
1197 |
+
"grad_norm": 0.09318208835708641,
|
1198 |
+
"learning_rate": 0.00015366406228495172,
|
1199 |
+
"loss": 0.341,
|
1200 |
+
"step": 169
|
1201 |
+
},
|
1202 |
+
{
|
1203 |
+
"epoch": 1.168384879725086,
|
1204 |
+
"grad_norm": 0.09374762758684776,
|
1205 |
+
"learning_rate": 0.00015298435524975572,
|
1206 |
+
"loss": 0.3336,
|
1207 |
+
"step": 170
|
1208 |
+
},
|
1209 |
+
{
|
1210 |
+
"epoch": 1.175257731958763,
|
1211 |
+
"grad_norm": 0.0941866112701508,
|
1212 |
+
"learning_rate": 0.00015230122769917527,
|
1213 |
+
"loss": 0.3391,
|
1214 |
+
"step": 171
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 1.1821305841924399,
|
1218 |
+
"grad_norm": 0.0911803650282977,
|
1219 |
+
"learning_rate": 0.00015161472373393186,
|
1220 |
+
"loss": 0.3322,
|
1221 |
+
"step": 172
|
1222 |
+
},
|
1223 |
+
{
|
1224 |
+
"epoch": 1.1890034364261168,
|
1225 |
+
"grad_norm": 0.09265526231900319,
|
1226 |
+
"learning_rate": 0.00015092488767271857,
|
1227 |
+
"loss": 0.3426,
|
1228 |
+
"step": 173
|
1229 |
+
},
|
1230 |
+
{
|
1231 |
+
"epoch": 1.1958762886597938,
|
1232 |
+
"grad_norm": 0.09286367954453642,
|
1233 |
+
"learning_rate": 0.00015023176404933874,
|
1234 |
+
"loss": 0.3456,
|
1235 |
+
"step": 174
|
1236 |
+
},
|
1237 |
+
{
|
1238 |
+
"epoch": 1.2027491408934707,
|
1239 |
+
"grad_norm": 0.08825029304085572,
|
1240 |
+
"learning_rate": 0.00014953539760983122,
|
1241 |
+
"loss": 0.3376,
|
1242 |
+
"step": 175
|
1243 |
+
},
|
1244 |
+
{
|
1245 |
+
"epoch": 1.2096219931271477,
|
1246 |
+
"grad_norm": 0.08762466707307968,
|
1247 |
+
"learning_rate": 0.0001488358333095816,
|
1248 |
+
"loss": 0.3294,
|
1249 |
+
"step": 176
|
1250 |
+
},
|
1251 |
+
{
|
1252 |
+
"epoch": 1.2164948453608249,
|
1253 |
+
"grad_norm": 0.08819497310411756,
|
1254 |
+
"learning_rate": 0.00014813311631041995,
|
1255 |
+
"loss": 0.3334,
|
1256 |
+
"step": 177
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.2233676975945018,
|
1260 |
+
"grad_norm": 0.08992362728208528,
|
1261 |
+
"learning_rate": 0.00014742729197770552,
|
1262 |
+
"loss": 0.3356,
|
1263 |
+
"step": 178
|
1264 |
+
},
|
1265 |
+
{
|
1266 |
+
"epoch": 1.2302405498281788,
|
1267 |
+
"grad_norm": 0.0884274642000476,
|
1268 |
+
"learning_rate": 0.00014671840587739783,
|
1269 |
+
"loss": 0.3362,
|
1270 |
+
"step": 179
|
1271 |
+
},
|
1272 |
+
{
|
1273 |
+
"epoch": 1.2371134020618557,
|
1274 |
+
"grad_norm": 0.0940460792061564,
|
1275 |
+
"learning_rate": 0.00014600650377311522,
|
1276 |
+
"loss": 0.3348,
|
1277 |
+
"step": 180
|
1278 |
+
},
|
1279 |
+
{
|
1280 |
+
"epoch": 1.2439862542955327,
|
1281 |
+
"grad_norm": 0.09439147324647236,
|
1282 |
+
"learning_rate": 0.0001452916316231805,
|
1283 |
+
"loss": 0.3393,
|
1284 |
+
"step": 181
|
1285 |
+
},
|
1286 |
+
{
|
1287 |
+
"epoch": 1.2508591065292096,
|
1288 |
+
"grad_norm": 0.08879817686299983,
|
1289 |
+
"learning_rate": 0.00014457383557765386,
|
1290 |
+
"loss": 0.3274,
|
1291 |
+
"step": 182
|
1292 |
+
},
|
1293 |
+
{
|
1294 |
+
"epoch": 1.2577319587628866,
|
1295 |
+
"grad_norm": 0.09128079327090716,
|
1296 |
+
"learning_rate": 0.00014385316197535372,
|
1297 |
+
"loss": 0.3445,
|
1298 |
+
"step": 183
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 1.2646048109965635,
|
1302 |
+
"grad_norm": 0.0885811126622494,
|
1303 |
+
"learning_rate": 0.00014312965734086518,
|
1304 |
+
"loss": 0.3377,
|
1305 |
+
"step": 184
|
1306 |
+
},
|
1307 |
+
{
|
1308 |
+
"epoch": 1.2714776632302405,
|
1309 |
+
"grad_norm": 0.08590906578126883,
|
1310 |
+
"learning_rate": 0.0001424033683815365,
|
1311 |
+
"loss": 0.3447,
|
1312 |
+
"step": 185
|
1313 |
+
},
|
1314 |
+
{
|
1315 |
+
"epoch": 1.2783505154639174,
|
1316 |
+
"grad_norm": 0.0935762976319064,
|
1317 |
+
"learning_rate": 0.00014167434198446383,
|
1318 |
+
"loss": 0.3302,
|
1319 |
+
"step": 186
|
1320 |
+
},
|
1321 |
+
{
|
1322 |
+
"epoch": 1.2852233676975944,
|
1323 |
+
"grad_norm": 0.0884675845519324,
|
1324 |
+
"learning_rate": 0.00014094262521346427,
|
1325 |
+
"loss": 0.3468,
|
1326 |
+
"step": 187
|
1327 |
+
},
|
1328 |
+
{
|
1329 |
+
"epoch": 1.2920962199312716,
|
1330 |
+
"grad_norm": 0.08665535174291548,
|
1331 |
+
"learning_rate": 0.00014020826530603776,
|
1332 |
+
"loss": 0.3339,
|
1333 |
+
"step": 188
|
1334 |
+
},
|
1335 |
+
{
|
1336 |
+
"epoch": 1.2989690721649485,
|
1337 |
+
"grad_norm": 0.08947173714380358,
|
1338 |
+
"learning_rate": 0.00013947130967031717,
|
1339 |
+
"loss": 0.3471,
|
1340 |
+
"step": 189
|
1341 |
+
},
|
1342 |
+
{
|
1343 |
+
"epoch": 1.3058419243986255,
|
1344 |
+
"grad_norm": 0.09103968023197583,
|
1345 |
+
"learning_rate": 0.00013873180588200827,
|
1346 |
+
"loss": 0.3412,
|
1347 |
+
"step": 190
|
1348 |
+
},
|
1349 |
+
{
|
1350 |
+
"epoch": 1.3127147766323024,
|
1351 |
+
"grad_norm": 0.0895152638091311,
|
1352 |
+
"learning_rate": 0.00013798980168131794,
|
1353 |
+
"loss": 0.3511,
|
1354 |
+
"step": 191
|
1355 |
+
},
|
1356 |
+
{
|
1357 |
+
"epoch": 1.3195876288659794,
|
1358 |
+
"grad_norm": 0.09024902028044313,
|
1359 |
+
"learning_rate": 0.00013724534496987247,
|
1360 |
+
"loss": 0.3352,
|
1361 |
+
"step": 192
|
1362 |
+
},
|
1363 |
+
{
|
1364 |
+
"epoch": 1.3264604810996563,
|
1365 |
+
"grad_norm": 0.09490877675864108,
|
1366 |
+
"learning_rate": 0.00013649848380762513,
|
1367 |
+
"loss": 0.3307,
|
1368 |
+
"step": 193
|
1369 |
+
},
|
1370 |
+
{
|
1371 |
+
"epoch": 1.3333333333333333,
|
1372 |
+
"grad_norm": 0.08952051838213773,
|
1373 |
+
"learning_rate": 0.0001357492664097534,
|
1374 |
+
"loss": 0.3337,
|
1375 |
+
"step": 194
|
1376 |
+
},
|
1377 |
+
{
|
1378 |
+
"epoch": 1.3402061855670104,
|
1379 |
+
"grad_norm": 0.089216758760593,
|
1380 |
+
"learning_rate": 0.00013499774114354655,
|
1381 |
+
"loss": 0.3401,
|
1382 |
+
"step": 195
|
1383 |
+
},
|
1384 |
+
{
|
1385 |
+
"epoch": 1.3470790378006874,
|
1386 |
+
"grad_norm": 0.09652202556469687,
|
1387 |
+
"learning_rate": 0.0001342439565252831,
|
1388 |
+
"loss": 0.3504,
|
1389 |
+
"step": 196
|
1390 |
+
},
|
1391 |
+
{
|
1392 |
+
"epoch": 1.3539518900343643,
|
1393 |
+
"grad_norm": 0.09323041932459158,
|
1394 |
+
"learning_rate": 0.00013348796121709862,
|
1395 |
+
"loss": 0.3519,
|
1396 |
+
"step": 197
|
1397 |
+
},
|
1398 |
+
{
|
1399 |
+
"epoch": 1.3608247422680413,
|
1400 |
+
"grad_norm": 0.08850056409502477,
|
1401 |
+
"learning_rate": 0.0001327298040238446,
|
1402 |
+
"loss": 0.3297,
|
1403 |
+
"step": 198
|
1404 |
+
},
|
1405 |
+
{
|
1406 |
+
"epoch": 1.3676975945017182,
|
1407 |
+
"grad_norm": 0.09367046869570124,
|
1408 |
+
"learning_rate": 0.00013196953388993726,
|
1409 |
+
"loss": 0.3401,
|
1410 |
+
"step": 199
|
1411 |
+
},
|
1412 |
+
{
|
1413 |
+
"epoch": 1.3745704467353952,
|
1414 |
+
"grad_norm": 0.08679754865679469,
|
1415 |
+
"learning_rate": 0.00013120719989619833,
|
1416 |
+
"loss": 0.3298,
|
1417 |
+
"step": 200
|
1418 |
+
},
|
1419 |
+
{
|
1420 |
+
"epoch": 1.3814432989690721,
|
1421 |
+
"grad_norm": 0.08896526634364879,
|
1422 |
+
"learning_rate": 0.00013044285125668614,
|
1423 |
+
"loss": 0.3278,
|
1424 |
+
"step": 201
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 1.388316151202749,
|
1428 |
+
"grad_norm": 0.09222733121952845,
|
1429 |
+
"learning_rate": 0.0001296765373155188,
|
1430 |
+
"loss": 0.3363,
|
1431 |
+
"step": 202
|
1432 |
+
},
|
1433 |
+
{
|
1434 |
+
"epoch": 1.395189003436426,
|
1435 |
+
"grad_norm": 0.08938179346059633,
|
1436 |
+
"learning_rate": 0.00012890830754368855,
|
1437 |
+
"loss": 0.3473,
|
1438 |
+
"step": 203
|
1439 |
+
},
|
1440 |
+
{
|
1441 |
+
"epoch": 1.402061855670103,
|
1442 |
+
"grad_norm": 0.08962836073074296,
|
1443 |
+
"learning_rate": 0.0001281382115358679,
|
1444 |
+
"loss": 0.3359,
|
1445 |
+
"step": 204
|
1446 |
+
},
|
1447 |
+
{
|
1448 |
+
"epoch": 1.40893470790378,
|
1449 |
+
"grad_norm": 0.08976699651731018,
|
1450 |
+
"learning_rate": 0.0001273662990072083,
|
1451 |
+
"loss": 0.3406,
|
1452 |
+
"step": 205
|
1453 |
+
},
|
1454 |
+
{
|
1455 |
+
"epoch": 1.4158075601374571,
|
1456 |
+
"grad_norm": 0.08730765725755779,
|
1457 |
+
"learning_rate": 0.00012659261979013043,
|
1458 |
+
"loss": 0.3261,
|
1459 |
+
"step": 206
|
1460 |
+
},
|
1461 |
+
{
|
1462 |
+
"epoch": 1.422680412371134,
|
1463 |
+
"grad_norm": 0.09242821980942448,
|
1464 |
+
"learning_rate": 0.00012581722383110718,
|
1465 |
+
"loss": 0.3323,
|
1466 |
+
"step": 207
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 1.429553264604811,
|
1470 |
+
"grad_norm": 0.08992082487698899,
|
1471 |
+
"learning_rate": 0.00012504016118743935,
|
1472 |
+
"loss": 0.3334,
|
1473 |
+
"step": 208
|
1474 |
+
},
|
1475 |
+
{
|
1476 |
+
"epoch": 1.436426116838488,
|
1477 |
+
"grad_norm": 0.09100065558206795,
|
1478 |
+
"learning_rate": 0.00012426148202402404,
|
1479 |
+
"loss": 0.3318,
|
1480 |
+
"step": 209
|
1481 |
+
},
|
1482 |
+
{
|
1483 |
+
"epoch": 1.443298969072165,
|
1484 |
+
"grad_norm": 0.0854247611013308,
|
1485 |
+
"learning_rate": 0.00012348123661011601,
|
1486 |
+
"loss": 0.3307,
|
1487 |
+
"step": 210
|
1488 |
+
},
|
1489 |
+
{
|
1490 |
+
"epoch": 1.4501718213058419,
|
1491 |
+
"grad_norm": 0.0865804470218671,
|
1492 |
+
"learning_rate": 0.00012269947531608276,
|
1493 |
+
"loss": 0.3507,
|
1494 |
+
"step": 211
|
1495 |
+
},
|
1496 |
+
{
|
1497 |
+
"epoch": 1.4570446735395188,
|
1498 |
+
"grad_norm": 0.09101321036566336,
|
1499 |
+
"learning_rate": 0.00012191624861015254,
|
1500 |
+
"loss": 0.3319,
|
1501 |
+
"step": 212
|
1502 |
+
},
|
1503 |
+
{
|
1504 |
+
"epoch": 1.463917525773196,
|
1505 |
+
"grad_norm": 0.08701002264437945,
|
1506 |
+
"learning_rate": 0.00012113160705515625,
|
1507 |
+
"loss": 0.333,
|
1508 |
+
"step": 213
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 1.470790378006873,
|
1512 |
+
"grad_norm": 0.08547913410680931,
|
1513 |
+
"learning_rate": 0.0001203456013052634,
|
1514 |
+
"loss": 0.3339,
|
1515 |
+
"step": 214
|
1516 |
+
},
|
1517 |
+
{
|
1518 |
+
"epoch": 1.47766323024055,
|
1519 |
+
"grad_norm": 0.08794901542532255,
|
1520 |
+
"learning_rate": 0.00011955828210271187,
|
1521 |
+
"loss": 0.3312,
|
1522 |
+
"step": 215
|
1523 |
+
},
|
1524 |
+
{
|
1525 |
+
"epoch": 1.4845360824742269,
|
1526 |
+
"grad_norm": 0.09333762500891976,
|
1527 |
+
"learning_rate": 0.00011876970027453222,
|
1528 |
+
"loss": 0.3306,
|
1529 |
+
"step": 216
|
1530 |
+
},
|
1531 |
+
{
|
1532 |
+
"epoch": 1.4914089347079038,
|
1533 |
+
"grad_norm": 0.09120345875867585,
|
1534 |
+
"learning_rate": 0.00011797990672926652,
|
1535 |
+
"loss": 0.3255,
|
1536 |
+
"step": 217
|
1537 |
+
},
|
1538 |
+
{
|
1539 |
+
"epoch": 1.4982817869415808,
|
1540 |
+
"grad_norm": 0.09150560929901355,
|
1541 |
+
"learning_rate": 0.00011718895245368167,
|
1542 |
+
"loss": 0.3291,
|
1543 |
+
"step": 218
|
1544 |
+
},
|
1545 |
+
{
|
1546 |
+
"epoch": 1.5051546391752577,
|
1547 |
+
"grad_norm": 0.09559054018841065,
|
1548 |
+
"learning_rate": 0.00011639688850947799,
|
1549 |
+
"loss": 0.3342,
|
1550 |
+
"step": 219
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 1.5120274914089347,
|
1554 |
+
"grad_norm": 0.0879293082122264,
|
1555 |
+
"learning_rate": 0.00011560376602999272,
|
1556 |
+
"loss": 0.3379,
|
1557 |
+
"step": 220
|
1558 |
+
},
|
1559 |
+
{
|
1560 |
+
"epoch": 1.5189003436426116,
|
1561 |
+
"grad_norm": 0.08863247791454076,
|
1562 |
+
"learning_rate": 0.00011480963621689905,
|
1563 |
+
"loss": 0.3295,
|
1564 |
+
"step": 221
|
1565 |
+
},
|
1566 |
+
{
|
1567 |
+
"epoch": 1.5257731958762886,
|
1568 |
+
"grad_norm": 0.088718267090734,
|
1569 |
+
"learning_rate": 0.00011401455033690076,
|
1570 |
+
"loss": 0.3253,
|
1571 |
+
"step": 222
|
1572 |
+
},
|
1573 |
+
{
|
1574 |
+
"epoch": 1.5326460481099655,
|
1575 |
+
"grad_norm": 0.0895731576057003,
|
1576 |
+
"learning_rate": 0.00011321855971842243,
|
1577 |
+
"loss": 0.3451,
|
1578 |
+
"step": 223
|
1579 |
+
},
|
1580 |
+
{
|
1581 |
+
"epoch": 1.5395189003436425,
|
1582 |
+
"grad_norm": 0.09120559405108837,
|
1583 |
+
"learning_rate": 0.00011242171574829599,
|
1584 |
+
"loss": 0.3267,
|
1585 |
+
"step": 224
|
1586 |
+
},
|
1587 |
+
{
|
1588 |
+
"epoch": 1.5463917525773194,
|
1589 |
+
"grad_norm": 0.08743574062739913,
|
1590 |
+
"learning_rate": 0.00011162406986844323,
|
1591 |
+
"loss": 0.3397,
|
1592 |
+
"step": 225
|
1593 |
+
},
|
1594 |
+
{
|
1595 |
+
"epoch": 1.5532646048109966,
|
1596 |
+
"grad_norm": 0.08824918438345017,
|
1597 |
+
"learning_rate": 0.00011082567357255484,
|
1598 |
+
"loss": 0.3328,
|
1599 |
+
"step": 226
|
1600 |
+
},
|
1601 |
+
{
|
1602 |
+
"epoch": 1.5601374570446735,
|
1603 |
+
"grad_norm": 0.090336802265287,
|
1604 |
+
"learning_rate": 0.00011002657840276627,
|
1605 |
+
"loss": 0.3269,
|
1606 |
+
"step": 227
|
1607 |
+
},
|
1608 |
+
{
|
1609 |
+
"epoch": 1.5670103092783505,
|
1610 |
+
"grad_norm": 0.09470893690130039,
|
1611 |
+
"learning_rate": 0.00010922683594633021,
|
1612 |
+
"loss": 0.3342,
|
1613 |
+
"step": 228
|
1614 |
+
},
|
1615 |
+
{
|
1616 |
+
"epoch": 1.5738831615120275,
|
1617 |
+
"grad_norm": 0.08950178767387156,
|
1618 |
+
"learning_rate": 0.00010842649783228624,
|
1619 |
+
"loss": 0.3361,
|
1620 |
+
"step": 229
|
1621 |
+
},
|
1622 |
+
{
|
1623 |
+
"epoch": 1.5807560137457046,
|
1624 |
+
"grad_norm": 0.09214111507393315,
|
1625 |
+
"learning_rate": 0.00010762561572812788,
|
1626 |
+
"loss": 0.3264,
|
1627 |
+
"step": 230
|
1628 |
+
},
|
1629 |
+
{
|
1630 |
+
"epoch": 1.5876288659793816,
|
1631 |
+
"grad_norm": 0.0935002910301003,
|
1632 |
+
"learning_rate": 0.0001068242413364671,
|
1633 |
+
"loss": 0.3255,
|
1634 |
+
"step": 231
|
1635 |
+
},
|
1636 |
+
{
|
1637 |
+
"epoch": 1.5945017182130585,
|
1638 |
+
"grad_norm": 0.0946979315371662,
|
1639 |
+
"learning_rate": 0.00010602242639169648,
|
1640 |
+
"loss": 0.322,
|
1641 |
+
"step": 232
|
1642 |
+
},
|
1643 |
+
{
|
1644 |
+
"epoch": 1.6013745704467355,
|
1645 |
+
"grad_norm": 0.0885334206141603,
|
1646 |
+
"learning_rate": 0.0001052202226566494,
|
1647 |
+
"loss": 0.3324,
|
1648 |
+
"step": 233
|
1649 |
+
},
|
1650 |
+
{
|
1651 |
+
"epoch": 1.6082474226804124,
|
1652 |
+
"grad_norm": 0.0907146677208864,
|
1653 |
+
"learning_rate": 0.00010441768191925847,
|
1654 |
+
"loss": 0.3301,
|
1655 |
+
"step": 234
|
1656 |
+
},
|
1657 |
+
{
|
1658 |
+
"epoch": 1.6151202749140894,
|
1659 |
+
"grad_norm": 0.0875343563564282,
|
1660 |
+
"learning_rate": 0.00010361485598921212,
|
1661 |
+
"loss": 0.3261,
|
1662 |
+
"step": 235
|
1663 |
+
},
|
1664 |
+
{
|
1665 |
+
"epoch": 1.6219931271477663,
|
1666 |
+
"grad_norm": 0.08662675484505016,
|
1667 |
+
"learning_rate": 0.00010281179669461005,
|
1668 |
+
"loss": 0.3364,
|
1669 |
+
"step": 236
|
1670 |
+
},
|
1671 |
+
{
|
1672 |
+
"epoch": 1.6288659793814433,
|
1673 |
+
"grad_norm": 0.08984998801891264,
|
1674 |
+
"learning_rate": 0.00010200855587861724,
|
1675 |
+
"loss": 0.33,
|
1676 |
+
"step": 237
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 1.6357388316151202,
|
1680 |
+
"grad_norm": 0.0896004179201218,
|
1681 |
+
"learning_rate": 0.0001012051853961172,
|
1682 |
+
"loss": 0.3302,
|
1683 |
+
"step": 238
|
1684 |
+
},
|
1685 |
+
{
|
1686 |
+
"epoch": 1.6426116838487972,
|
1687 |
+
"grad_norm": 0.09055262757285264,
|
1688 |
+
"learning_rate": 0.00010040173711036431,
|
1689 |
+
"loss": 0.3388,
|
1690 |
+
"step": 239
|
1691 |
+
},
|
1692 |
+
{
|
1693 |
+
"epoch": 1.6494845360824741,
|
1694 |
+
"grad_norm": 0.08453899569938862,
|
1695 |
+
"learning_rate": 9.959826288963571e-05,
|
1696 |
+
"loss": 0.3256,
|
1697 |
+
"step": 240
|
1698 |
+
},
|
1699 |
+
{
|
1700 |
+
"epoch": 1.656357388316151,
|
1701 |
+
"grad_norm": 0.0889936265286851,
|
1702 |
+
"learning_rate": 9.879481460388282e-05,
|
1703 |
+
"loss": 0.3253,
|
1704 |
+
"step": 241
|
1705 |
+
},
|
1706 |
+
{
|
1707 |
+
"epoch": 1.663230240549828,
|
1708 |
+
"grad_norm": 0.09422158003783734,
|
1709 |
+
"learning_rate": 9.799144412138275e-05,
|
1710 |
+
"loss": 0.3395,
|
1711 |
+
"step": 242
|
1712 |
+
},
|
1713 |
+
{
|
1714 |
+
"epoch": 1.670103092783505,
|
1715 |
+
"grad_norm": 0.09145977121685474,
|
1716 |
+
"learning_rate": 9.718820330538998e-05,
|
1717 |
+
"loss": 0.3399,
|
1718 |
+
"step": 243
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 1.6769759450171822,
|
1722 |
+
"grad_norm": 0.08738667677613643,
|
1723 |
+
"learning_rate": 9.638514401078788e-05,
|
1724 |
+
"loss": 0.3216,
|
1725 |
+
"step": 244
|
1726 |
+
},
|
1727 |
+
{
|
1728 |
+
"epoch": 1.6838487972508591,
|
1729 |
+
"grad_norm": 0.088092907094314,
|
1730 |
+
"learning_rate": 9.558231808074156e-05,
|
1731 |
+
"loss": 0.3364,
|
1732 |
+
"step": 245
|
1733 |
+
},
|
1734 |
+
{
|
1735 |
+
"epoch": 1.690721649484536,
|
1736 |
+
"grad_norm": 0.09109969761880952,
|
1737 |
+
"learning_rate": 9.477977734335061e-05,
|
1738 |
+
"loss": 0.3149,
|
1739 |
+
"step": 246
|
1740 |
+
},
|
1741 |
+
{
|
1742 |
+
"epoch": 1.697594501718213,
|
1743 |
+
"grad_norm": 0.08963068066688211,
|
1744 |
+
"learning_rate": 9.397757360830353e-05,
|
1745 |
+
"loss": 0.3294,
|
1746 |
+
"step": 247
|
1747 |
+
},
|
1748 |
+
{
|
1749 |
+
"epoch": 1.7044673539518902,
|
1750 |
+
"grad_norm": 0.09658512621034114,
|
1751 |
+
"learning_rate": 9.317575866353292e-05,
|
1752 |
+
"loss": 0.3358,
|
1753 |
+
"step": 248
|
1754 |
+
},
|
1755 |
+
{
|
1756 |
+
"epoch": 1.7113402061855671,
|
1757 |
+
"grad_norm": 0.088750132481195,
|
1758 |
+
"learning_rate": 9.23743842718721e-05,
|
1759 |
+
"loss": 0.3361,
|
1760 |
+
"step": 249
|
1761 |
+
},
|
1762 |
+
{
|
1763 |
+
"epoch": 1.718213058419244,
|
1764 |
+
"grad_norm": 0.08920547189763123,
|
1765 |
+
"learning_rate": 9.157350216771378e-05,
|
1766 |
+
"loss": 0.3407,
|
1767 |
+
"step": 250
|
1768 |
+
},
|
1769 |
+
{
|
1770 |
+
"epoch": 1.725085910652921,
|
1771 |
+
"grad_norm": 0.08974488907293028,
|
1772 |
+
"learning_rate": 9.077316405366981e-05,
|
1773 |
+
"loss": 0.3278,
|
1774 |
+
"step": 251
|
1775 |
+
},
|
1776 |
+
{
|
1777 |
+
"epoch": 1.731958762886598,
|
1778 |
+
"grad_norm": 0.08706782434665275,
|
1779 |
+
"learning_rate": 8.997342159723371e-05,
|
1780 |
+
"loss": 0.3307,
|
1781 |
+
"step": 252
|
1782 |
+
},
|
1783 |
+
{
|
1784 |
+
"epoch": 1.738831615120275,
|
1785 |
+
"grad_norm": 0.08783198274350164,
|
1786 |
+
"learning_rate": 8.917432642744518e-05,
|
1787 |
+
"loss": 0.3308,
|
1788 |
+
"step": 253
|
1789 |
+
},
|
1790 |
+
{
|
1791 |
+
"epoch": 1.745704467353952,
|
1792 |
+
"grad_norm": 0.09166745552607744,
|
1793 |
+
"learning_rate": 8.83759301315568e-05,
|
1794 |
+
"loss": 0.3157,
|
1795 |
+
"step": 254
|
1796 |
+
},
|
1797 |
+
{
|
1798 |
+
"epoch": 1.7525773195876289,
|
1799 |
+
"grad_norm": 0.0908320442953557,
|
1800 |
+
"learning_rate": 8.757828425170404e-05,
|
1801 |
+
"loss": 0.3199,
|
1802 |
+
"step": 255
|
1803 |
+
},
|
1804 |
+
{
|
1805 |
+
"epoch": 1.7594501718213058,
|
1806 |
+
"grad_norm": 0.0849446544243993,
|
1807 |
+
"learning_rate": 8.678144028157759e-05,
|
1808 |
+
"loss": 0.3228,
|
1809 |
+
"step": 256
|
1810 |
+
},
|
1811 |
+
{
|
1812 |
+
"epoch": 1.7663230240549828,
|
1813 |
+
"grad_norm": 0.08521177018484068,
|
1814 |
+
"learning_rate": 8.598544966309925e-05,
|
1815 |
+
"loss": 0.3338,
|
1816 |
+
"step": 257
|
1817 |
+
},
|
1818 |
+
{
|
1819 |
+
"epoch": 1.7731958762886597,
|
1820 |
+
"grad_norm": 0.08387408489546401,
|
1821 |
+
"learning_rate": 8.519036378310096e-05,
|
1822 |
+
"loss": 0.3116,
|
1823 |
+
"step": 258
|
1824 |
+
},
|
1825 |
+
{
|
1826 |
+
"epoch": 1.7800687285223367,
|
1827 |
+
"grad_norm": 0.08864985384052881,
|
1828 |
+
"learning_rate": 8.43962339700073e-05,
|
1829 |
+
"loss": 0.3324,
|
1830 |
+
"step": 259
|
1831 |
+
},
|
1832 |
+
{
|
1833 |
+
"epoch": 1.7869415807560136,
|
1834 |
+
"grad_norm": 0.08641650021361377,
|
1835 |
+
"learning_rate": 8.360311149052205e-05,
|
1836 |
+
"loss": 0.3372,
|
1837 |
+
"step": 260
|
1838 |
+
},
|
1839 |
+
{
|
1840 |
+
"epoch": 1.7938144329896906,
|
1841 |
+
"grad_norm": 0.08706064763076157,
|
1842 |
+
"learning_rate": 8.281104754631835e-05,
|
1843 |
+
"loss": 0.3289,
|
1844 |
+
"step": 261
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 1.8006872852233677,
|
1848 |
+
"grad_norm": 0.08847890936614504,
|
1849 |
+
"learning_rate": 8.20200932707335e-05,
|
1850 |
+
"loss": 0.3168,
|
1851 |
+
"step": 262
|
1852 |
+
},
|
1853 |
+
{
|
1854 |
+
"epoch": 1.8075601374570447,
|
1855 |
+
"grad_norm": 0.08453149563340354,
|
1856 |
+
"learning_rate": 8.123029972546781e-05,
|
1857 |
+
"loss": 0.3259,
|
1858 |
+
"step": 263
|
1859 |
+
},
|
1860 |
+
{
|
1861 |
+
"epoch": 1.8144329896907216,
|
1862 |
+
"grad_norm": 0.08479101242481418,
|
1863 |
+
"learning_rate": 8.044171789728816e-05,
|
1864 |
+
"loss": 0.3358,
|
1865 |
+
"step": 264
|
1866 |
+
},
|
1867 |
+
{
|
1868 |
+
"epoch": 1.8213058419243986,
|
1869 |
+
"grad_norm": 0.08899658479991718,
|
1870 |
+
"learning_rate": 7.965439869473664e-05,
|
1871 |
+
"loss": 0.3391,
|
1872 |
+
"step": 265
|
1873 |
+
},
|
1874 |
+
{
|
1875 |
+
"epoch": 1.8281786941580758,
|
1876 |
+
"grad_norm": 0.0847681466880256,
|
1877 |
+
"learning_rate": 7.886839294484377e-05,
|
1878 |
+
"loss": 0.3207,
|
1879 |
+
"step": 266
|
1880 |
+
},
|
1881 |
+
{
|
1882 |
+
"epoch": 1.8350515463917527,
|
1883 |
+
"grad_norm": 0.08677934509062958,
|
1884 |
+
"learning_rate": 7.808375138984745e-05,
|
1885 |
+
"loss": 0.3365,
|
1886 |
+
"step": 267
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 1.8419243986254297,
|
1890 |
+
"grad_norm": 0.08816011534386732,
|
1891 |
+
"learning_rate": 7.730052468391725e-05,
|
1892 |
+
"loss": 0.3366,
|
1893 |
+
"step": 268
|
1894 |
+
},
|
1895 |
+
{
|
1896 |
+
"epoch": 1.8487972508591066,
|
1897 |
+
"grad_norm": 0.08456107767474944,
|
1898 |
+
"learning_rate": 7.6518763389884e-05,
|
1899 |
+
"loss": 0.3053,
|
1900 |
+
"step": 269
|
1901 |
+
},
|
1902 |
+
{
|
1903 |
+
"epoch": 1.8556701030927836,
|
1904 |
+
"grad_norm": 0.08664098772150312,
|
1905 |
+
"learning_rate": 7.573851797597602e-05,
|
1906 |
+
"loss": 0.3142,
|
1907 |
+
"step": 270
|
1908 |
+
},
|
1909 |
+
{
|
1910 |
+
"epoch": 1.8625429553264605,
|
1911 |
+
"grad_norm": 0.08474629067803426,
|
1912 |
+
"learning_rate": 7.495983881256067e-05,
|
1913 |
+
"loss": 0.3286,
|
1914 |
+
"step": 271
|
1915 |
+
},
|
1916 |
+
{
|
1917 |
+
"epoch": 1.8694158075601375,
|
1918 |
+
"grad_norm": 0.09338120416229846,
|
1919 |
+
"learning_rate": 7.418277616889282e-05,
|
1920 |
+
"loss": 0.3411,
|
1921 |
+
"step": 272
|
1922 |
+
},
|
1923 |
+
{
|
1924 |
+
"epoch": 1.8762886597938144,
|
1925 |
+
"grad_norm": 0.09029648849822869,
|
1926 |
+
"learning_rate": 7.340738020986961e-05,
|
1927 |
+
"loss": 0.3104,
|
1928 |
+
"step": 273
|
1929 |
+
},
|
1930 |
+
{
|
1931 |
+
"epoch": 1.8831615120274914,
|
1932 |
+
"grad_norm": 0.09059323374606333,
|
1933 |
+
"learning_rate": 7.263370099279172e-05,
|
1934 |
+
"loss": 0.3242,
|
1935 |
+
"step": 274
|
1936 |
+
},
|
1937 |
+
{
|
1938 |
+
"epoch": 1.8900343642611683,
|
1939 |
+
"grad_norm": 0.08762544318436638,
|
1940 |
+
"learning_rate": 7.186178846413214e-05,
|
1941 |
+
"loss": 0.3264,
|
1942 |
+
"step": 275
|
1943 |
+
},
|
1944 |
+
{
|
1945 |
+
"epoch": 1.8969072164948453,
|
1946 |
+
"grad_norm": 0.09069509247176193,
|
1947 |
+
"learning_rate": 7.109169245631149e-05,
|
1948 |
+
"loss": 0.3434,
|
1949 |
+
"step": 276
|
1950 |
+
},
|
1951 |
+
{
|
1952 |
+
"epoch": 1.9037800687285222,
|
1953 |
+
"grad_norm": 0.09065127457729591,
|
1954 |
+
"learning_rate": 7.032346268448118e-05,
|
1955 |
+
"loss": 0.3259,
|
1956 |
+
"step": 277
|
1957 |
+
},
|
1958 |
+
{
|
1959 |
+
"epoch": 1.9106529209621992,
|
1960 |
+
"grad_norm": 0.09060438836672054,
|
1961 |
+
"learning_rate": 6.955714874331387e-05,
|
1962 |
+
"loss": 0.3285,
|
1963 |
+
"step": 278
|
1964 |
+
},
|
1965 |
+
{
|
1966 |
+
"epoch": 1.9175257731958761,
|
1967 |
+
"grad_norm": 0.08937375112557829,
|
1968 |
+
"learning_rate": 6.87928001038017e-05,
|
1969 |
+
"loss": 0.3297,
|
1970 |
+
"step": 279
|
1971 |
+
},
|
1972 |
+
{
|
1973 |
+
"epoch": 1.9243986254295533,
|
1974 |
+
"grad_norm": 0.08590914133335634,
|
1975 |
+
"learning_rate": 6.803046611006278e-05,
|
1976 |
+
"loss": 0.3318,
|
1977 |
+
"step": 280
|
1978 |
+
},
|
1979 |
+
{
|
1980 |
+
"epoch": 1.9312714776632303,
|
1981 |
+
"grad_norm": 0.08712353618570348,
|
1982 |
+
"learning_rate": 6.727019597615545e-05,
|
1983 |
+
"loss": 0.3357,
|
1984 |
+
"step": 281
|
1985 |
+
},
|
1986 |
+
{
|
1987 |
+
"epoch": 1.9381443298969072,
|
1988 |
+
"grad_norm": 0.09015854175252304,
|
1989 |
+
"learning_rate": 6.651203878290139e-05,
|
1990 |
+
"loss": 0.33,
|
1991 |
+
"step": 282
|
1992 |
+
},
|
1993 |
+
{
|
1994 |
+
"epoch": 1.9450171821305842,
|
1995 |
+
"grad_norm": 0.0866982085957751,
|
1996 |
+
"learning_rate": 6.575604347471695e-05,
|
1997 |
+
"loss": 0.3342,
|
1998 |
+
"step": 283
|
1999 |
+
},
|
2000 |
+
{
|
2001 |
+
"epoch": 1.9518900343642611,
|
2002 |
+
"grad_norm": 0.08675769846482406,
|
2003 |
+
"learning_rate": 6.500225885645346e-05,
|
2004 |
+
"loss": 0.328,
|
2005 |
+
"step": 284
|
2006 |
+
},
|
2007 |
+
{
|
2008 |
+
"epoch": 1.9587628865979383,
|
2009 |
+
"grad_norm": 0.08889126479207589,
|
2010 |
+
"learning_rate": 6.425073359024663e-05,
|
2011 |
+
"loss": 0.3322,
|
2012 |
+
"step": 285
|
2013 |
+
},
|
2014 |
+
{
|
2015 |
+
"epoch": 1.9656357388316152,
|
2016 |
+
"grad_norm": 0.08989017017161777,
|
2017 |
+
"learning_rate": 6.350151619237488e-05,
|
2018 |
+
"loss": 0.3192,
|
2019 |
+
"step": 286
|
2020 |
+
},
|
2021 |
+
{
|
2022 |
+
"epoch": 1.9725085910652922,
|
2023 |
+
"grad_norm": 0.08930847261254722,
|
2024 |
+
"learning_rate": 6.275465503012751e-05,
|
2025 |
+
"loss": 0.3282,
|
2026 |
+
"step": 287
|
2027 |
+
},
|
2028 |
+
{
|
2029 |
+
"epoch": 1.9793814432989691,
|
2030 |
+
"grad_norm": 0.08954697412786164,
|
2031 |
+
"learning_rate": 6.201019831868208e-05,
|
2032 |
+
"loss": 0.3282,
|
2033 |
+
"step": 288
|
2034 |
+
},
|
2035 |
+
{
|
2036 |
+
"epoch": 1.986254295532646,
|
2037 |
+
"grad_norm": 0.0893723471550625,
|
2038 |
+
"learning_rate": 6.126819411799175e-05,
|
2039 |
+
"loss": 0.3258,
|
2040 |
+
"step": 289
|
2041 |
+
},
|
2042 |
+
{
|
2043 |
+
"epoch": 1.993127147766323,
|
2044 |
+
"grad_norm": 0.09367805430596919,
|
2045 |
+
"learning_rate": 6.052869032968285e-05,
|
2046 |
+
"loss": 0.333,
|
2047 |
+
"step": 290
|
2048 |
+
},
|
2049 |
+
{
|
2050 |
+
"epoch": 2.0,
|
2051 |
+
"grad_norm": 0.0918567679510941,
|
2052 |
+
"learning_rate": 5.979173469396227e-05,
|
2053 |
+
"loss": 0.3258,
|
2054 |
+
"step": 291
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 2.0,
|
2058 |
+
"eval_loss": 0.339150071144104,
|
2059 |
+
"eval_runtime": 29.8937,
|
2060 |
+
"eval_samples_per_second": 32.749,
|
2061 |
+
"eval_steps_per_second": 1.037,
|
2062 |
+
"step": 291
|
2063 |
+
},
|
2064 |
+
{
|
2065 |
+
"epoch": 2.006872852233677,
|
2066 |
+
"grad_norm": 0.0886467750539302,
|
2067 |
+
"learning_rate": 5.905737478653572e-05,
|
2068 |
+
"loss": 0.3202,
|
2069 |
+
"step": 292
|
2070 |
+
},
|
2071 |
+
{
|
2072 |
+
"epoch": 2.013745704467354,
|
2073 |
+
"grad_norm": 0.08711918167350534,
|
2074 |
+
"learning_rate": 5.83256580155362e-05,
|
2075 |
+
"loss": 0.3083,
|
2076 |
+
"step": 293
|
2077 |
+
},
|
2078 |
+
{
|
2079 |
+
"epoch": 2.020618556701031,
|
2080 |
+
"grad_norm": 0.08819504515806559,
|
2081 |
+
"learning_rate": 5.7596631618463514e-05,
|
2082 |
+
"loss": 0.3038,
|
2083 |
+
"step": 294
|
2084 |
+
},
|
2085 |
+
{
|
2086 |
+
"epoch": 2.027491408934708,
|
2087 |
+
"grad_norm": 0.08581578620028243,
|
2088 |
+
"learning_rate": 5.687034265913485e-05,
|
2089 |
+
"loss": 0.309,
|
2090 |
+
"step": 295
|
2091 |
+
},
|
2092 |
+
{
|
2093 |
+
"epoch": 2.0343642611683848,
|
2094 |
+
"grad_norm": 0.08813058468520787,
|
2095 |
+
"learning_rate": 5.614683802464631e-05,
|
2096 |
+
"loss": 0.309,
|
2097 |
+
"step": 296
|
2098 |
+
},
|
2099 |
+
{
|
2100 |
+
"epoch": 2.0412371134020617,
|
2101 |
+
"grad_norm": 0.0916296719651795,
|
2102 |
+
"learning_rate": 5.542616442234618e-05,
|
2103 |
+
"loss": 0.3095,
|
2104 |
+
"step": 297
|
2105 |
+
},
|
2106 |
+
{
|
2107 |
+
"epoch": 2.0481099656357387,
|
2108 |
+
"grad_norm": 0.09093120997153643,
|
2109 |
+
"learning_rate": 5.470836837681954e-05,
|
2110 |
+
"loss": 0.3086,
|
2111 |
+
"step": 298
|
2112 |
+
},
|
2113 |
+
{
|
2114 |
+
"epoch": 2.0549828178694156,
|
2115 |
+
"grad_norm": 0.09273763656782547,
|
2116 |
+
"learning_rate": 5.399349622688479e-05,
|
2117 |
+
"loss": 0.3061,
|
2118 |
+
"step": 299
|
2119 |
+
},
|
2120 |
+
{
|
2121 |
+
"epoch": 2.0618556701030926,
|
2122 |
+
"grad_norm": 0.09545728182493257,
|
2123 |
+
"learning_rate": 5.32815941226022e-05,
|
2124 |
+
"loss": 0.3146,
|
2125 |
+
"step": 300
|
2126 |
+
},
|
2127 |
+
{
|
2128 |
+
"epoch": 2.06872852233677,
|
2129 |
+
"grad_norm": 0.0948791991068102,
|
2130 |
+
"learning_rate": 5.2572708022294504e-05,
|
2131 |
+
"loss": 0.3131,
|
2132 |
+
"step": 301
|
2133 |
+
},
|
2134 |
+
{
|
2135 |
+
"epoch": 2.075601374570447,
|
2136 |
+
"grad_norm": 0.0892496983847854,
|
2137 |
+
"learning_rate": 5.1866883689580056e-05,
|
2138 |
+
"loss": 0.3031,
|
2139 |
+
"step": 302
|
2140 |
+
},
|
2141 |
+
{
|
2142 |
+
"epoch": 2.082474226804124,
|
2143 |
+
"grad_norm": 0.09114623555977093,
|
2144 |
+
"learning_rate": 5.116416669041843e-05,
|
2145 |
+
"loss": 0.3208,
|
2146 |
+
"step": 303
|
2147 |
+
},
|
2148 |
+
{
|
2149 |
+
"epoch": 2.089347079037801,
|
2150 |
+
"grad_norm": 0.08920660312195167,
|
2151 |
+
"learning_rate": 5.046460239016879e-05,
|
2152 |
+
"loss": 0.3051,
|
2153 |
+
"step": 304
|
2154 |
+
},
|
2155 |
+
{
|
2156 |
+
"epoch": 2.0962199312714778,
|
2157 |
+
"grad_norm": 0.08984052643157422,
|
2158 |
+
"learning_rate": 4.976823595066128e-05,
|
2159 |
+
"loss": 0.3172,
|
2160 |
+
"step": 305
|
2161 |
+
},
|
2162 |
+
{
|
2163 |
+
"epoch": 2.1030927835051547,
|
2164 |
+
"grad_norm": 0.09466282282051863,
|
2165 |
+
"learning_rate": 4.907511232728145e-05,
|
2166 |
+
"loss": 0.3198,
|
2167 |
+
"step": 306
|
2168 |
+
},
|
2169 |
+
{
|
2170 |
+
"epoch": 2.1099656357388317,
|
2171 |
+
"grad_norm": 0.09231441594067813,
|
2172 |
+
"learning_rate": 4.8385276266068146e-05,
|
2173 |
+
"loss": 0.3099,
|
2174 |
+
"step": 307
|
2175 |
+
},
|
2176 |
+
{
|
2177 |
+
"epoch": 2.1168384879725086,
|
2178 |
+
"grad_norm": 0.09260548846821774,
|
2179 |
+
"learning_rate": 4.7698772300824756e-05,
|
2180 |
+
"loss": 0.3039,
|
2181 |
+
"step": 308
|
2182 |
+
},
|
2183 |
+
{
|
2184 |
+
"epoch": 2.1237113402061856,
|
2185 |
+
"grad_norm": 0.09326606205682852,
|
2186 |
+
"learning_rate": 4.7015644750244306e-05,
|
2187 |
+
"loss": 0.312,
|
2188 |
+
"step": 309
|
2189 |
+
},
|
2190 |
+
{
|
2191 |
+
"epoch": 2.1305841924398625,
|
2192 |
+
"grad_norm": 0.0948368176467578,
|
2193 |
+
"learning_rate": 4.6335937715048306e-05,
|
2194 |
+
"loss": 0.3101,
|
2195 |
+
"step": 310
|
2196 |
+
},
|
2197 |
+
{
|
2198 |
+
"epoch": 2.1374570446735395,
|
2199 |
+
"grad_norm": 0.09530056231484316,
|
2200 |
+
"learning_rate": 4.565969507513981e-05,
|
2201 |
+
"loss": 0.317,
|
2202 |
+
"step": 311
|
2203 |
+
},
|
2204 |
+
{
|
2205 |
+
"epoch": 2.1443298969072164,
|
2206 |
+
"grad_norm": 0.08767980465231293,
|
2207 |
+
"learning_rate": 4.498696048677059e-05,
|
2208 |
+
"loss": 0.2959,
|
2209 |
+
"step": 312
|
2210 |
+
},
|
2211 |
+
{
|
2212 |
+
"epoch": 2.1512027491408934,
|
2213 |
+
"grad_norm": 0.09563893318480989,
|
2214 |
+
"learning_rate": 4.4317777379722866e-05,
|
2215 |
+
"loss": 0.3097,
|
2216 |
+
"step": 313
|
2217 |
+
},
|
2218 |
+
{
|
2219 |
+
"epoch": 2.1580756013745703,
|
2220 |
+
"grad_norm": 0.09383833971767949,
|
2221 |
+
"learning_rate": 4.365218895450558e-05,
|
2222 |
+
"loss": 0.3185,
|
2223 |
+
"step": 314
|
2224 |
+
},
|
2225 |
+
{
|
2226 |
+
"epoch": 2.1649484536082473,
|
2227 |
+
"grad_norm": 0.0939267618670719,
|
2228 |
+
"learning_rate": 4.29902381795655e-05,
|
2229 |
+
"loss": 0.3145,
|
2230 |
+
"step": 315
|
2231 |
+
},
|
2232 |
+
{
|
2233 |
+
"epoch": 2.1718213058419242,
|
2234 |
+
"grad_norm": 0.09389687242828468,
|
2235 |
+
"learning_rate": 4.2331967788513295e-05,
|
2236 |
+
"loss": 0.3071,
|
2237 |
+
"step": 316
|
2238 |
+
},
|
2239 |
+
{
|
2240 |
+
"epoch": 2.178694158075601,
|
2241 |
+
"grad_norm": 0.09487597194212076,
|
2242 |
+
"learning_rate": 4.167742027736482e-05,
|
2243 |
+
"loss": 0.3193,
|
2244 |
+
"step": 317
|
2245 |
+
},
|
2246 |
+
{
|
2247 |
+
"epoch": 2.1855670103092786,
|
2248 |
+
"grad_norm": 0.0935473777981694,
|
2249 |
+
"learning_rate": 4.102663790179764e-05,
|
2250 |
+
"loss": 0.3118,
|
2251 |
+
"step": 318
|
2252 |
+
},
|
2253 |
+
{
|
2254 |
+
"epoch": 2.1924398625429555,
|
2255 |
+
"grad_norm": 0.09393015139691238,
|
2256 |
+
"learning_rate": 4.037966267442315e-05,
|
2257 |
+
"loss": 0.3044,
|
2258 |
+
"step": 319
|
2259 |
+
},
|
2260 |
+
{
|
2261 |
+
"epoch": 2.1993127147766325,
|
2262 |
+
"grad_norm": 0.09265614548574853,
|
2263 |
+
"learning_rate": 3.973653636207437e-05,
|
2264 |
+
"loss": 0.3221,
|
2265 |
+
"step": 320
|
2266 |
+
},
|
2267 |
+
{
|
2268 |
+
"epoch": 2.2061855670103094,
|
2269 |
+
"grad_norm": 0.09322822287056864,
|
2270 |
+
"learning_rate": 3.909730048310962e-05,
|
2271 |
+
"loss": 0.3102,
|
2272 |
+
"step": 321
|
2273 |
+
},
|
2274 |
+
{
|
2275 |
+
"epoch": 2.2130584192439864,
|
2276 |
+
"grad_norm": 0.08955741861746863,
|
2277 |
+
"learning_rate": 3.846199630473216e-05,
|
2278 |
+
"loss": 0.3118,
|
2279 |
+
"step": 322
|
2280 |
+
},
|
2281 |
+
{
|
2282 |
+
"epoch": 2.2199312714776633,
|
2283 |
+
"grad_norm": 0.08967712595094215,
|
2284 |
+
"learning_rate": 3.7830664840326145e-05,
|
2285 |
+
"loss": 0.2978,
|
2286 |
+
"step": 323
|
2287 |
+
},
|
2288 |
+
{
|
2289 |
+
"epoch": 2.2268041237113403,
|
2290 |
+
"grad_norm": 0.09178434696992077,
|
2291 |
+
"learning_rate": 3.720334684680889e-05,
|
2292 |
+
"loss": 0.3069,
|
2293 |
+
"step": 324
|
2294 |
+
},
|
2295 |
+
{
|
2296 |
+
"epoch": 2.2336769759450172,
|
2297 |
+
"grad_norm": 0.09152559857353068,
|
2298 |
+
"learning_rate": 3.6580082821999786e-05,
|
2299 |
+
"loss": 0.3044,
|
2300 |
+
"step": 325
|
2301 |
+
},
|
2302 |
+
{
|
2303 |
+
"epoch": 2.240549828178694,
|
2304 |
+
"grad_norm": 0.09594190960981773,
|
2305 |
+
"learning_rate": 3.596091300200578e-05,
|
2306 |
+
"loss": 0.3044,
|
2307 |
+
"step": 326
|
2308 |
+
},
|
2309 |
+
{
|
2310 |
+
"epoch": 2.247422680412371,
|
2311 |
+
"grad_norm": 0.0926964681182163,
|
2312 |
+
"learning_rate": 3.534587735862391e-05,
|
2313 |
+
"loss": 0.2969,
|
2314 |
+
"step": 327
|
2315 |
+
},
|
2316 |
+
{
|
2317 |
+
"epoch": 2.254295532646048,
|
2318 |
+
"grad_norm": 0.09701468065123808,
|
2319 |
+
"learning_rate": 3.473501559676088e-05,
|
2320 |
+
"loss": 0.3235,
|
2321 |
+
"step": 328
|
2322 |
+
},
|
2323 |
+
{
|
2324 |
+
"epoch": 2.261168384879725,
|
2325 |
+
"grad_norm": 0.09133515228348349,
|
2326 |
+
"learning_rate": 3.4128367151869714e-05,
|
2327 |
+
"loss": 0.298,
|
2328 |
+
"step": 329
|
2329 |
+
},
|
2330 |
+
{
|
2331 |
+
"epoch": 2.268041237113402,
|
2332 |
+
"grad_norm": 0.08947864352930586,
|
2333 |
+
"learning_rate": 3.352597118740404e-05,
|
2334 |
+
"loss": 0.2991,
|
2335 |
+
"step": 330
|
2336 |
+
},
|
2337 |
+
{
|
2338 |
+
"epoch": 2.274914089347079,
|
2339 |
+
"grad_norm": 0.09297738601504922,
|
2340 |
+
"learning_rate": 3.292786659228973e-05,
|
2341 |
+
"loss": 0.3068,
|
2342 |
+
"step": 331
|
2343 |
+
},
|
2344 |
+
{
|
2345 |
+
"epoch": 2.281786941580756,
|
2346 |
+
"grad_norm": 0.08884667943821722,
|
2347 |
+
"learning_rate": 3.233409197841437e-05,
|
2348 |
+
"loss": 0.3052,
|
2349 |
+
"step": 332
|
2350 |
+
},
|
2351 |
+
{
|
2352 |
+
"epoch": 2.288659793814433,
|
2353 |
+
"grad_norm": 0.0899494873645269,
|
2354 |
+
"learning_rate": 3.174468567813461e-05,
|
2355 |
+
"loss": 0.3081,
|
2356 |
+
"step": 333
|
2357 |
+
},
|
2358 |
+
{
|
2359 |
+
"epoch": 2.29553264604811,
|
2360 |
+
"grad_norm": 0.09543617597300157,
|
2361 |
+
"learning_rate": 3.115968574180149e-05,
|
2362 |
+
"loss": 0.3009,
|
2363 |
+
"step": 334
|
2364 |
+
},
|
2365 |
+
{
|
2366 |
+
"epoch": 2.3024054982817868,
|
2367 |
+
"grad_norm": 0.09901153510644121,
|
2368 |
+
"learning_rate": 3.0579129935304066e-05,
|
2369 |
+
"loss": 0.3102,
|
2370 |
+
"step": 335
|
2371 |
+
},
|
2372 |
+
{
|
2373 |
+
"epoch": 2.3092783505154637,
|
2374 |
+
"grad_norm": 0.08924299716097303,
|
2375 |
+
"learning_rate": 3.0003055737631403e-05,
|
2376 |
+
"loss": 0.3034,
|
2377 |
+
"step": 336
|
2378 |
+
},
|
2379 |
+
{
|
2380 |
+
"epoch": 2.3161512027491407,
|
2381 |
+
"grad_norm": 0.09482046351771685,
|
2382 |
+
"learning_rate": 2.9431500338452832e-05,
|
2383 |
+
"loss": 0.3093,
|
2384 |
+
"step": 337
|
2385 |
+
},
|
2386 |
+
{
|
2387 |
+
"epoch": 2.323024054982818,
|
2388 |
+
"grad_norm": 0.09205312585551002,
|
2389 |
+
"learning_rate": 2.886450063571735e-05,
|
2390 |
+
"loss": 0.2946,
|
2391 |
+
"step": 338
|
2392 |
+
},
|
2393 |
+
{
|
2394 |
+
"epoch": 2.329896907216495,
|
2395 |
+
"grad_norm": 0.09331131336969856,
|
2396 |
+
"learning_rate": 2.8302093233271453e-05,
|
2397 |
+
"loss": 0.2984,
|
2398 |
+
"step": 339
|
2399 |
+
},
|
2400 |
+
{
|
2401 |
+
"epoch": 2.336769759450172,
|
2402 |
+
"grad_norm": 0.09491376569079615,
|
2403 |
+
"learning_rate": 2.7744314438496088e-05,
|
2404 |
+
"loss": 0.31,
|
2405 |
+
"step": 340
|
2406 |
+
},
|
2407 |
+
{
|
2408 |
+
"epoch": 2.343642611683849,
|
2409 |
+
"grad_norm": 0.0895175338196936,
|
2410 |
+
"learning_rate": 2.7191200259962934e-05,
|
2411 |
+
"loss": 0.3044,
|
2412 |
+
"step": 341
|
2413 |
+
},
|
2414 |
+
{
|
2415 |
+
"epoch": 2.350515463917526,
|
2416 |
+
"grad_norm": 0.0927041757839274,
|
2417 |
+
"learning_rate": 2.6642786405109475e-05,
|
2418 |
+
"loss": 0.3078,
|
2419 |
+
"step": 342
|
2420 |
+
},
|
2421 |
+
{
|
2422 |
+
"epoch": 2.357388316151203,
|
2423 |
+
"grad_norm": 0.09534749181408379,
|
2424 |
+
"learning_rate": 2.6099108277934103e-05,
|
2425 |
+
"loss": 0.3026,
|
2426 |
+
"step": 343
|
2427 |
+
},
|
2428 |
+
{
|
2429 |
+
"epoch": 2.3642611683848798,
|
2430 |
+
"grad_norm": 0.09275326056910925,
|
2431 |
+
"learning_rate": 2.556020097671046e-05,
|
2432 |
+
"loss": 0.3027,
|
2433 |
+
"step": 344
|
2434 |
+
},
|
2435 |
+
{
|
2436 |
+
"epoch": 2.3711340206185567,
|
2437 |
+
"grad_norm": 0.09223933442175777,
|
2438 |
+
"learning_rate": 2.5026099291721516e-05,
|
2439 |
+
"loss": 0.304,
|
2440 |
+
"step": 345
|
2441 |
+
},
|
2442 |
+
{
|
2443 |
+
"epoch": 2.3780068728522337,
|
2444 |
+
"grad_norm": 0.09068677016558731,
|
2445 |
+
"learning_rate": 2.449683770301382e-05,
|
2446 |
+
"loss": 0.2999,
|
2447 |
+
"step": 346
|
2448 |
+
},
|
2449 |
+
{
|
2450 |
+
"epoch": 2.3848797250859106,
|
2451 |
+
"grad_norm": 0.09958738181261272,
|
2452 |
+
"learning_rate": 2.397245037817125e-05,
|
2453 |
+
"loss": 0.3142,
|
2454 |
+
"step": 347
|
2455 |
+
},
|
2456 |
+
{
|
2457 |
+
"epoch": 2.3917525773195876,
|
2458 |
+
"grad_norm": 0.09717094561100138,
|
2459 |
+
"learning_rate": 2.345297117010954e-05,
|
2460 |
+
"loss": 0.3051,
|
2461 |
+
"step": 348
|
2462 |
+
},
|
2463 |
+
{
|
2464 |
+
"epoch": 2.3986254295532645,
|
2465 |
+
"grad_norm": 0.09182595410158129,
|
2466 |
+
"learning_rate": 2.2938433614890697e-05,
|
2467 |
+
"loss": 0.3025,
|
2468 |
+
"step": 349
|
2469 |
+
},
|
2470 |
+
{
|
2471 |
+
"epoch": 2.4054982817869415,
|
2472 |
+
"grad_norm": 0.0977647503565851,
|
2473 |
+
"learning_rate": 2.242887092955801e-05,
|
2474 |
+
"loss": 0.3204,
|
2475 |
+
"step": 350
|
2476 |
+
},
|
2477 |
+
{
|
2478 |
+
"epoch": 2.4123711340206184,
|
2479 |
+
"grad_norm": 0.09283949853643803,
|
2480 |
+
"learning_rate": 2.1924316009991787e-05,
|
2481 |
+
"loss": 0.3158,
|
2482 |
+
"step": 351
|
2483 |
+
},
|
2484 |
+
{
|
2485 |
+
"epoch": 2.4192439862542954,
|
2486 |
+
"grad_norm": 0.09566793807194976,
|
2487 |
+
"learning_rate": 2.1424801428785447e-05,
|
2488 |
+
"loss": 0.3152,
|
2489 |
+
"step": 352
|
2490 |
+
},
|
2491 |
+
{
|
2492 |
+
"epoch": 2.4261168384879723,
|
2493 |
+
"grad_norm": 0.09359864147013991,
|
2494 |
+
"learning_rate": 2.0930359433142932e-05,
|
2495 |
+
"loss": 0.3112,
|
2496 |
+
"step": 353
|
2497 |
+
},
|
2498 |
+
{
|
2499 |
+
"epoch": 2.4329896907216497,
|
2500 |
+
"grad_norm": 0.09298886578820206,
|
2501 |
+
"learning_rate": 2.0441021942796944e-05,
|
2502 |
+
"loss": 0.3063,
|
2503 |
+
"step": 354
|
2504 |
+
},
|
2505 |
+
{
|
2506 |
+
"epoch": 2.4398625429553267,
|
2507 |
+
"grad_norm": 0.08950379993278665,
|
2508 |
+
"learning_rate": 1.995682054794803e-05,
|
2509 |
+
"loss": 0.2985,
|
2510 |
+
"step": 355
|
2511 |
+
},
|
2512 |
+
{
|
2513 |
+
"epoch": 2.4467353951890036,
|
2514 |
+
"grad_norm": 0.0931691133735062,
|
2515 |
+
"learning_rate": 1.9477786507225616e-05,
|
2516 |
+
"loss": 0.2951,
|
2517 |
+
"step": 356
|
2518 |
+
},
|
2519 |
+
{
|
2520 |
+
"epoch": 2.4536082474226806,
|
2521 |
+
"grad_norm": 0.09472765677973273,
|
2522 |
+
"learning_rate": 1.900395074566962e-05,
|
2523 |
+
"loss": 0.3126,
|
2524 |
+
"step": 357
|
2525 |
+
},
|
2526 |
+
{
|
2527 |
+
"epoch": 2.4604810996563575,
|
2528 |
+
"grad_norm": 0.09573566390044937,
|
2529 |
+
"learning_rate": 1.8535343852734332e-05,
|
2530 |
+
"loss": 0.3179,
|
2531 |
+
"step": 358
|
2532 |
+
},
|
2533 |
+
{
|
2534 |
+
"epoch": 2.4673539518900345,
|
2535 |
+
"grad_norm": 0.09271039293485658,
|
2536 |
+
"learning_rate": 1.8071996080313602e-05,
|
2537 |
+
"loss": 0.3073,
|
2538 |
+
"step": 359
|
2539 |
+
},
|
2540 |
+
{
|
2541 |
+
"epoch": 2.4742268041237114,
|
2542 |
+
"grad_norm": 0.09540958215019993,
|
2543 |
+
"learning_rate": 1.76139373407876e-05,
|
2544 |
+
"loss": 0.3026,
|
2545 |
+
"step": 360
|
2546 |
+
},
|
2547 |
+
{
|
2548 |
+
"epoch": 2.4810996563573884,
|
2549 |
+
"grad_norm": 0.09425927677933707,
|
2550 |
+
"learning_rate": 1.7161197205092216e-05,
|
2551 |
+
"loss": 0.3097,
|
2552 |
+
"step": 361
|
2553 |
+
},
|
2554 |
+
{
|
2555 |
+
"epoch": 2.4879725085910653,
|
2556 |
+
"grad_norm": 0.09238356291671689,
|
2557 |
+
"learning_rate": 1.6713804900809582e-05,
|
2558 |
+
"loss": 0.321,
|
2559 |
+
"step": 362
|
2560 |
+
},
|
2561 |
+
{
|
2562 |
+
"epoch": 2.4948453608247423,
|
2563 |
+
"grad_norm": 0.09134364572635581,
|
2564 |
+
"learning_rate": 1.6271789310281517e-05,
|
2565 |
+
"loss": 0.3227,
|
2566 |
+
"step": 363
|
2567 |
+
},
|
2568 |
+
{
|
2569 |
+
"epoch": 2.5017182130584192,
|
2570 |
+
"grad_norm": 0.09593922917122714,
|
2571 |
+
"learning_rate": 1.583517896874498e-05,
|
2572 |
+
"loss": 0.307,
|
2573 |
+
"step": 364
|
2574 |
+
},
|
2575 |
+
{
|
2576 |
+
"epoch": 2.508591065292096,
|
2577 |
+
"grad_norm": 0.09535245685017667,
|
2578 |
+
"learning_rate": 1.540400206248963e-05,
|
2579 |
+
"loss": 0.3064,
|
2580 |
+
"step": 365
|
2581 |
+
},
|
2582 |
+
{
|
2583 |
+
"epoch": 2.515463917525773,
|
2584 |
+
"grad_norm": 0.09696139813613523,
|
2585 |
+
"learning_rate": 1.4978286427038601e-05,
|
2586 |
+
"loss": 0.3094,
|
2587 |
+
"step": 366
|
2588 |
+
},
|
2589 |
+
{
|
2590 |
+
"epoch": 2.52233676975945,
|
2591 |
+
"grad_norm": 0.0949873193903831,
|
2592 |
+
"learning_rate": 1.4558059545351143e-05,
|
2593 |
+
"loss": 0.3122,
|
2594 |
+
"step": 367
|
2595 |
+
},
|
2596 |
+
{
|
2597 |
+
"epoch": 2.529209621993127,
|
2598 |
+
"grad_norm": 0.09354418491130151,
|
2599 |
+
"learning_rate": 1.4143348546048707e-05,
|
2600 |
+
"loss": 0.3063,
|
2601 |
+
"step": 368
|
2602 |
+
},
|
2603 |
+
{
|
2604 |
+
"epoch": 2.536082474226804,
|
2605 |
+
"grad_norm": 0.09290641840755422,
|
2606 |
+
"learning_rate": 1.3734180201663439e-05,
|
2607 |
+
"loss": 0.312,
|
2608 |
+
"step": 369
|
2609 |
+
},
|
2610 |
+
{
|
2611 |
+
"epoch": 2.542955326460481,
|
2612 |
+
"grad_norm": 0.09174814409428318,
|
2613 |
+
"learning_rate": 1.3330580926909763e-05,
|
2614 |
+
"loss": 0.3044,
|
2615 |
+
"step": 370
|
2616 |
+
},
|
2617 |
+
{
|
2618 |
+
"epoch": 2.549828178694158,
|
2619 |
+
"grad_norm": 0.09549880298792861,
|
2620 |
+
"learning_rate": 1.2932576776979377e-05,
|
2621 |
+
"loss": 0.3198,
|
2622 |
+
"step": 371
|
2623 |
+
},
|
2624 |
+
{
|
2625 |
+
"epoch": 2.556701030927835,
|
2626 |
+
"grad_norm": 0.09244963446933405,
|
2627 |
+
"learning_rate": 1.2540193445858883e-05,
|
2628 |
+
"loss": 0.3125,
|
2629 |
+
"step": 372
|
2630 |
+
},
|
2631 |
+
{
|
2632 |
+
"epoch": 2.563573883161512,
|
2633 |
+
"grad_norm": 0.09344004805774578,
|
2634 |
+
"learning_rate": 1.2153456264671337e-05,
|
2635 |
+
"loss": 0.3023,
|
2636 |
+
"step": 373
|
2637 |
+
},
|
2638 |
+
{
|
2639 |
+
"epoch": 2.5704467353951888,
|
2640 |
+
"grad_norm": 0.09184517524908066,
|
2641 |
+
"learning_rate": 1.1772390200040817e-05,
|
2642 |
+
"loss": 0.2936,
|
2643 |
+
"step": 374
|
2644 |
+
},
|
2645 |
+
{
|
2646 |
+
"epoch": 2.5773195876288657,
|
2647 |
+
"grad_norm": 0.09342477534524925,
|
2648 |
+
"learning_rate": 1.139701985248055e-05,
|
2649 |
+
"loss": 0.3119,
|
2650 |
+
"step": 375
|
2651 |
+
},
|
2652 |
+
{
|
2653 |
+
"epoch": 2.584192439862543,
|
2654 |
+
"grad_norm": 0.09179966838370937,
|
2655 |
+
"learning_rate": 1.1027369454805058e-05,
|
2656 |
+
"loss": 0.3043,
|
2657 |
+
"step": 376
|
2658 |
+
},
|
2659 |
+
{
|
2660 |
+
"epoch": 2.59106529209622,
|
2661 |
+
"grad_norm": 0.09202317210863532,
|
2662 |
+
"learning_rate": 1.0663462870565411e-05,
|
2663 |
+
"loss": 0.3151,
|
2664 |
+
"step": 377
|
2665 |
+
},
|
2666 |
+
{
|
2667 |
+
"epoch": 2.597938144329897,
|
2668 |
+
"grad_norm": 0.09393910730229943,
|
2669 |
+
"learning_rate": 1.0305323592509009e-05,
|
2670 |
+
"loss": 0.3123,
|
2671 |
+
"step": 378
|
2672 |
+
},
|
2673 |
+
{
|
2674 |
+
"epoch": 2.604810996563574,
|
2675 |
+
"grad_norm": 0.09493170283217371,
|
2676 |
+
"learning_rate": 9.952974741062703e-06,
|
2677 |
+
"loss": 0.3033,
|
2678 |
+
"step": 379
|
2679 |
+
},
|
2680 |
+
{
|
2681 |
+
"epoch": 2.611683848797251,
|
2682 |
+
"grad_norm": 0.09299379885008295,
|
2683 |
+
"learning_rate": 9.606439062840256e-06,
|
2684 |
+
"loss": 0.3058,
|
2685 |
+
"step": 380
|
2686 |
+
},
|
2687 |
+
{
|
2688 |
+
"epoch": 2.618556701030928,
|
2689 |
+
"grad_norm": 0.09583678306809282,
|
2690 |
+
"learning_rate": 9.265738929174051e-06,
|
2691 |
+
"loss": 0.3006,
|
2692 |
+
"step": 381
|
2693 |
+
},
|
2694 |
+
{
|
2695 |
+
"epoch": 2.625429553264605,
|
2696 |
+
"grad_norm": 0.09387939203068772,
|
2697 |
+
"learning_rate": 8.93089633467058e-06,
|
2698 |
+
"loss": 0.3,
|
2699 |
+
"step": 382
|
2700 |
+
},
|
2701 |
+
{
|
2702 |
+
"epoch": 2.6323024054982818,
|
2703 |
+
"grad_norm": 0.09499675770463291,
|
2704 |
+
"learning_rate": 8.601932895790877e-06,
|
2705 |
+
"loss": 0.2997,
|
2706 |
+
"step": 383
|
2707 |
+
},
|
2708 |
+
{
|
2709 |
+
"epoch": 2.6391752577319587,
|
2710 |
+
"grad_norm": 0.09410604007614645,
|
2711 |
+
"learning_rate": 8.278869849454718e-06,
|
2712 |
+
"loss": 0.3037,
|
2713 |
+
"step": 384
|
2714 |
+
},
|
2715 |
+
{
|
2716 |
+
"epoch": 2.6460481099656357,
|
2717 |
+
"grad_norm": 0.09190526072343619,
|
2718 |
+
"learning_rate": 7.961728051669737e-06,
|
2719 |
+
"loss": 0.309,
|
2720 |
+
"step": 385
|
2721 |
+
},
|
2722 |
+
{
|
2723 |
+
"epoch": 2.6529209621993126,
|
2724 |
+
"grad_norm": 0.09556333829479592,
|
2725 |
+
"learning_rate": 7.650527976185173e-06,
|
2726 |
+
"loss": 0.3059,
|
2727 |
+
"step": 386
|
2728 |
+
},
|
2729 |
+
{
|
2730 |
+
"epoch": 2.6597938144329896,
|
2731 |
+
"grad_norm": 0.09577450734899226,
|
2732 |
+
"learning_rate": 7.3452897131698564e-06,
|
2733 |
+
"loss": 0.3111,
|
2734 |
+
"step": 387
|
2735 |
+
},
|
2736 |
+
{
|
2737 |
+
"epoch": 2.6666666666666665,
|
2738 |
+
"grad_norm": 0.09084914727337236,
|
2739 |
+
"learning_rate": 7.046032967915483e-06,
|
2740 |
+
"loss": 0.3202,
|
2741 |
+
"step": 388
|
2742 |
+
},
|
2743 |
+
{
|
2744 |
+
"epoch": 2.673539518900344,
|
2745 |
+
"grad_norm": 0.09500372962694525,
|
2746 |
+
"learning_rate": 6.75277705956443e-06,
|
2747 |
+
"loss": 0.3128,
|
2748 |
+
"step": 389
|
2749 |
+
},
|
2750 |
+
{
|
2751 |
+
"epoch": 2.680412371134021,
|
2752 |
+
"grad_norm": 0.09004758102796465,
|
2753 |
+
"learning_rate": 6.465540919862456e-06,
|
2754 |
+
"loss": 0.3015,
|
2755 |
+
"step": 390
|
2756 |
+
},
|
2757 |
+
{
|
2758 |
+
"epoch": 2.687285223367698,
|
2759 |
+
"grad_norm": 0.09131460491333845,
|
2760 |
+
"learning_rate": 6.184343091936751e-06,
|
2761 |
+
"loss": 0.3106,
|
2762 |
+
"step": 391
|
2763 |
+
},
|
2764 |
+
{
|
2765 |
+
"epoch": 2.6941580756013748,
|
2766 |
+
"grad_norm": 0.09299567382728216,
|
2767 |
+
"learning_rate": 5.909201729098579e-06,
|
2768 |
+
"loss": 0.3017,
|
2769 |
+
"step": 392
|
2770 |
+
},
|
2771 |
+
{
|
2772 |
+
"epoch": 2.7010309278350517,
|
2773 |
+
"grad_norm": 0.09429778610425,
|
2774 |
+
"learning_rate": 5.640134593671598e-06,
|
2775 |
+
"loss": 0.3015,
|
2776 |
+
"step": 393
|
2777 |
+
},
|
2778 |
+
{
|
2779 |
+
"epoch": 2.7079037800687287,
|
2780 |
+
"grad_norm": 0.09097995580479669,
|
2781 |
+
"learning_rate": 5.3771590558450265e-06,
|
2782 |
+
"loss": 0.2945,
|
2783 |
+
"step": 394
|
2784 |
+
},
|
2785 |
+
{
|
2786 |
+
"epoch": 2.7147766323024056,
|
2787 |
+
"grad_norm": 0.09448453183988398,
|
2788 |
+
"learning_rate": 5.12029209255227e-06,
|
2789 |
+
"loss": 0.3066,
|
2790 |
+
"step": 395
|
2791 |
+
},
|
2792 |
+
{
|
2793 |
+
"epoch": 2.7216494845360826,
|
2794 |
+
"grad_norm": 0.0951928424174606,
|
2795 |
+
"learning_rate": 4.869550286375091e-06,
|
2796 |
+
"loss": 0.3115,
|
2797 |
+
"step": 396
|
2798 |
+
},
|
2799 |
+
{
|
2800 |
+
"epoch": 2.7285223367697595,
|
2801 |
+
"grad_norm": 0.09301501151406104,
|
2802 |
+
"learning_rate": 4.624949824472858e-06,
|
2803 |
+
"loss": 0.3051,
|
2804 |
+
"step": 397
|
2805 |
+
},
|
2806 |
+
{
|
2807 |
+
"epoch": 2.7353951890034365,
|
2808 |
+
"grad_norm": 0.09407884130575386,
|
2809 |
+
"learning_rate": 4.386506497537757e-06,
|
2810 |
+
"loss": 0.295,
|
2811 |
+
"step": 398
|
2812 |
+
},
|
2813 |
+
{
|
2814 |
+
"epoch": 2.7422680412371134,
|
2815 |
+
"grad_norm": 0.0916280626690068,
|
2816 |
+
"learning_rate": 4.154235698775277e-06,
|
2817 |
+
"loss": 0.2939,
|
2818 |
+
"step": 399
|
2819 |
+
},
|
2820 |
+
{
|
2821 |
+
"epoch": 2.7491408934707904,
|
2822 |
+
"grad_norm": 0.09229074966846487,
|
2823 |
+
"learning_rate": 3.928152422910491e-06,
|
2824 |
+
"loss": 0.3031,
|
2825 |
+
"step": 400
|
2826 |
+
},
|
2827 |
+
{
|
2828 |
+
"epoch": 2.7560137457044673,
|
2829 |
+
"grad_norm": 0.0918136228014016,
|
2830 |
+
"learning_rate": 3.7082712652200867e-06,
|
2831 |
+
"loss": 0.2951,
|
2832 |
+
"step": 401
|
2833 |
+
},
|
2834 |
+
{
|
2835 |
+
"epoch": 2.7628865979381443,
|
2836 |
+
"grad_norm": 0.09132120510205273,
|
2837 |
+
"learning_rate": 3.4946064205899965e-06,
|
2838 |
+
"loss": 0.2898,
|
2839 |
+
"step": 402
|
2840 |
+
},
|
2841 |
+
{
|
2842 |
+
"epoch": 2.7697594501718212,
|
2843 |
+
"grad_norm": 0.09262224139833831,
|
2844 |
+
"learning_rate": 3.287171682599255e-06,
|
2845 |
+
"loss": 0.3037,
|
2846 |
+
"step": 403
|
2847 |
+
},
|
2848 |
+
{
|
2849 |
+
"epoch": 2.776632302405498,
|
2850 |
+
"grad_norm": 0.09384566411627011,
|
2851 |
+
"learning_rate": 3.085980442629288e-06,
|
2852 |
+
"loss": 0.305,
|
2853 |
+
"step": 404
|
2854 |
+
},
|
2855 |
+
{
|
2856 |
+
"epoch": 2.783505154639175,
|
2857 |
+
"grad_norm": 0.09699044066640952,
|
2858 |
+
"learning_rate": 2.8910456889995498e-06,
|
2859 |
+
"loss": 0.3124,
|
2860 |
+
"step": 405
|
2861 |
+
},
|
2862 |
+
{
|
2863 |
+
"epoch": 2.790378006872852,
|
2864 |
+
"grad_norm": 0.0929330783626954,
|
2865 |
+
"learning_rate": 2.7023800061289907e-06,
|
2866 |
+
"loss": 0.3014,
|
2867 |
+
"step": 406
|
2868 |
+
},
|
2869 |
+
{
|
2870 |
+
"epoch": 2.797250859106529,
|
2871 |
+
"grad_norm": 0.09098829237433458,
|
2872 |
+
"learning_rate": 2.5199955737236104e-06,
|
2873 |
+
"loss": 0.3063,
|
2874 |
+
"step": 407
|
2875 |
+
},
|
2876 |
+
{
|
2877 |
+
"epoch": 2.804123711340206,
|
2878 |
+
"grad_norm": 0.09219645759358688,
|
2879 |
+
"learning_rate": 2.3439041659902407e-06,
|
2880 |
+
"loss": 0.2993,
|
2881 |
+
"step": 408
|
2882 |
+
},
|
2883 |
+
{
|
2884 |
+
"epoch": 2.810996563573883,
|
2885 |
+
"grad_norm": 0.08981813230207816,
|
2886 |
+
"learning_rate": 2.174117150876398e-06,
|
2887 |
+
"loss": 0.2958,
|
2888 |
+
"step": 409
|
2889 |
+
},
|
2890 |
+
{
|
2891 |
+
"epoch": 2.81786941580756,
|
2892 |
+
"grad_norm": 0.0901341358982541,
|
2893 |
+
"learning_rate": 2.010645489336382e-06,
|
2894 |
+
"loss": 0.3075,
|
2895 |
+
"step": 410
|
2896 |
+
},
|
2897 |
+
{
|
2898 |
+
"epoch": 2.824742268041237,
|
2899 |
+
"grad_norm": 0.092777307675848,
|
2900 |
+
"learning_rate": 1.8534997346237093e-06,
|
2901 |
+
"loss": 0.299,
|
2902 |
+
"step": 411
|
2903 |
+
},
|
2904 |
+
{
|
2905 |
+
"epoch": 2.8316151202749142,
|
2906 |
+
"grad_norm": 0.09506315255714337,
|
2907 |
+
"learning_rate": 1.7026900316098215e-06,
|
2908 |
+
"loss": 0.301,
|
2909 |
+
"step": 412
|
2910 |
+
},
|
2911 |
+
{
|
2912 |
+
"epoch": 2.838487972508591,
|
2913 |
+
"grad_norm": 0.09223555543912211,
|
2914 |
+
"learning_rate": 1.5582261161291245e-06,
|
2915 |
+
"loss": 0.3067,
|
2916 |
+
"step": 413
|
2917 |
+
},
|
2918 |
+
{
|
2919 |
+
"epoch": 2.845360824742268,
|
2920 |
+
"grad_norm": 0.09520376073895358,
|
2921 |
+
"learning_rate": 1.4201173143504888e-06,
|
2922 |
+
"loss": 0.3103,
|
2923 |
+
"step": 414
|
2924 |
+
},
|
2925 |
+
{
|
2926 |
+
"epoch": 2.852233676975945,
|
2927 |
+
"grad_norm": 0.09303971646126957,
|
2928 |
+
"learning_rate": 1.2883725421752201e-06,
|
2929 |
+
"loss": 0.3069,
|
2930 |
+
"step": 415
|
2931 |
+
},
|
2932 |
+
{
|
2933 |
+
"epoch": 2.859106529209622,
|
2934 |
+
"grad_norm": 0.09437919113312288,
|
2935 |
+
"learning_rate": 1.1630003046614323e-06,
|
2936 |
+
"loss": 0.3035,
|
2937 |
+
"step": 416
|
2938 |
+
},
|
2939 |
+
{
|
2940 |
+
"epoch": 2.865979381443299,
|
2941 |
+
"grad_norm": 0.0942033239779668,
|
2942 |
+
"learning_rate": 1.0440086954749517e-06,
|
2943 |
+
"loss": 0.3132,
|
2944 |
+
"step": 417
|
2945 |
+
},
|
2946 |
+
{
|
2947 |
+
"epoch": 2.872852233676976,
|
2948 |
+
"grad_norm": 0.09554927057554605,
|
2949 |
+
"learning_rate": 9.314053963669245e-07,
|
2950 |
+
"loss": 0.3142,
|
2951 |
+
"step": 418
|
2952 |
+
},
|
2953 |
+
{
|
2954 |
+
"epoch": 2.879725085910653,
|
2955 |
+
"grad_norm": 0.09353322239406023,
|
2956 |
+
"learning_rate": 8.251976766777913e-07,
|
2957 |
+
"loss": 0.3106,
|
2958 |
+
"step": 419
|
2959 |
+
},
|
2960 |
+
{
|
2961 |
+
"epoch": 2.88659793814433,
|
2962 |
+
"grad_norm": 0.09300605711009886,
|
2963 |
+
"learning_rate": 7.253923928680406e-07,
|
2964 |
+
"loss": 0.3166,
|
2965 |
+
"step": 420
|
2966 |
+
},
|
2967 |
+
{
|
2968 |
+
"epoch": 2.893470790378007,
|
2969 |
+
"grad_norm": 0.09521857944477934,
|
2970 |
+
"learning_rate": 6.319959880756177e-07,
|
2971 |
+
"loss": 0.3097,
|
2972 |
+
"step": 421
|
2973 |
+
},
|
2974 |
+
{
|
2975 |
+
"epoch": 2.9003436426116838,
|
2976 |
+
"grad_norm": 0.09536928122104008,
|
2977 |
+
"learning_rate": 5.450144916999134e-07,
|
2978 |
+
"loss": 0.2955,
|
2979 |
+
"step": 422
|
2980 |
+
},
|
2981 |
+
{
|
2982 |
+
"epoch": 2.9072164948453607,
|
2983 |
+
"grad_norm": 0.09339292910298186,
|
2984 |
+
"learning_rate": 4.644535190125421e-07,
|
2985 |
+
"loss": 0.2997,
|
2986 |
+
"step": 423
|
2987 |
+
},
|
2988 |
+
{
|
2989 |
+
"epoch": 2.9140893470790377,
|
2990 |
+
"grad_norm": 0.09285035939013274,
|
2991 |
+
"learning_rate": 3.903182707948649e-07,
|
2992 |
+
"loss": 0.3034,
|
2993 |
+
"step": 424
|
2994 |
+
},
|
2995 |
+
{
|
2996 |
+
"epoch": 2.9209621993127146,
|
2997 |
+
"grad_norm": 0.09605475715998495,
|
2998 |
+
"learning_rate": 3.2261353300219176e-07,
|
2999 |
+
"loss": 0.3117,
|
3000 |
+
"step": 425
|
3001 |
+
},
|
3002 |
+
{
|
3003 |
+
"epoch": 2.927835051546392,
|
3004 |
+
"grad_norm": 0.09487799919270845,
|
3005 |
+
"learning_rate": 2.613436764548505e-07,
|
3006 |
+
"loss": 0.3106,
|
3007 |
+
"step": 426
|
3008 |
+
},
|
3009 |
+
{
|
3010 |
+
"epoch": 2.934707903780069,
|
3011 |
+
"grad_norm": 0.09654737369397567,
|
3012 |
+
"learning_rate": 2.0651265655603492e-07,
|
3013 |
+
"loss": 0.318,
|
3014 |
+
"step": 427
|
3015 |
+
},
|
3016 |
+
{
|
3017 |
+
"epoch": 2.941580756013746,
|
3018 |
+
"grad_norm": 0.09558890253304671,
|
3019 |
+
"learning_rate": 1.5812401303639813e-07,
|
3020 |
+
"loss": 0.3027,
|
3021 |
+
"step": 428
|
3022 |
+
},
|
3023 |
+
{
|
3024 |
+
"epoch": 2.948453608247423,
|
3025 |
+
"grad_norm": 0.09135373502935519,
|
3026 |
+
"learning_rate": 1.1618086972559062e-07,
|
3027 |
+
"loss": 0.3042,
|
3028 |
+
"step": 429
|
3029 |
+
},
|
3030 |
+
{
|
3031 |
+
"epoch": 2.9553264604811,
|
3032 |
+
"grad_norm": 0.09279803401969587,
|
3033 |
+
"learning_rate": 8.068593435055505e-08,
|
3034 |
+
"loss": 0.3,
|
3035 |
+
"step": 430
|
3036 |
+
},
|
3037 |
+
{
|
3038 |
+
"epoch": 2.9621993127147768,
|
3039 |
+
"grad_norm": 0.08748562231091243,
|
3040 |
+
"learning_rate": 5.164149836077714e-08,
|
3041 |
+
"loss": 0.2925,
|
3042 |
+
"step": 431
|
3043 |
+
},
|
3044 |
+
{
|
3045 |
+
"epoch": 2.9690721649484537,
|
3046 |
+
"grad_norm": 0.09421083042315682,
|
3047 |
+
"learning_rate": 2.9049436780281825e-08,
|
3048 |
+
"loss": 0.3194,
|
3049 |
+
"step": 432
|
3050 |
+
},
|
3051 |
+
{
|
3052 |
+
"epoch": 2.9759450171821307,
|
3053 |
+
"grad_norm": 0.0915890606460084,
|
3054 |
+
"learning_rate": 1.2911208086663351e-08,
|
3055 |
+
"loss": 0.2976,
|
3056 |
+
"step": 433
|
3057 |
+
},
|
3058 |
+
{
|
3059 |
+
"epoch": 2.9828178694158076,
|
3060 |
+
"grad_norm": 0.09059761210406012,
|
3061 |
+
"learning_rate": 3.2278541168717646e-09,
|
3062 |
+
"loss": 0.2912,
|
3063 |
+
"step": 434
|
3064 |
+
},
|
3065 |
+
{
|
3066 |
+
"epoch": 2.9896907216494846,
|
3067 |
+
"grad_norm": 0.09094828082389766,
|
3068 |
+
"learning_rate": 0.0,
|
3069 |
+
"loss": 0.3019,
|
3070 |
+
"step": 435
|
3071 |
+
},
|
3072 |
+
{
|
3073 |
+
"epoch": 2.9896907216494846,
|
3074 |
+
"eval_loss": 0.3361770212650299,
|
3075 |
+
"eval_runtime": 29.844,
|
3076 |
+
"eval_samples_per_second": 32.804,
|
3077 |
+
"eval_steps_per_second": 1.039,
|
3078 |
+
"step": 435
|
3079 |
+
},
|
3080 |
+
{
|
3081 |
+
"epoch": 2.9896907216494846,
|
3082 |
+
"step": 435,
|
3083 |
+
"total_flos": 1.566236469816197e+17,
|
3084 |
+
"train_loss": 0.3695115421695271,
|
3085 |
+
"train_runtime": 4901.7941,
|
3086 |
+
"train_samples_per_second": 11.382,
|
3087 |
+
"train_steps_per_second": 0.089
|
3088 |
+
}
|
3089 |
+
],
|
3090 |
+
"logging_steps": 1,
|
3091 |
+
"max_steps": 435,
|
3092 |
+
"num_input_tokens_seen": 0,
|
3093 |
+
"num_train_epochs": 3,
|
3094 |
+
"save_steps": 500,
|
3095 |
+
"stateful_callbacks": {
|
3096 |
+
"TrainerControl": {
|
3097 |
+
"args": {
|
3098 |
+
"should_epoch_stop": false,
|
3099 |
+
"should_evaluate": false,
|
3100 |
+
"should_log": false,
|
3101 |
+
"should_save": true,
|
3102 |
+
"should_training_stop": true
|
3103 |
+
},
|
3104 |
+
"attributes": {}
|
3105 |
+
}
|
3106 |
+
},
|
3107 |
+
"total_flos": 1.566236469816197e+17,
|
3108 |
+
"train_batch_size": 8,
|
3109 |
+
"trial_name": null,
|
3110 |
+
"trial_params": null
|
3111 |
+
}
|