Model save
Browse files- README.md +69 -0
- all_results.json +9 -0
- runs/Sep23_22-52-56_2e80285b3fe9/events.out.tfevents.1727132349.2e80285b3fe9.120709.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2248 -0
README.md
ADDED
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model: google/gemma-7b
|
3 |
+
datasets:
|
4 |
+
- generator
|
5 |
+
library_name: peft
|
6 |
+
license: gemma
|
7 |
+
tags:
|
8 |
+
- trl
|
9 |
+
- sft
|
10 |
+
- generated_from_trainer
|
11 |
+
model-index:
|
12 |
+
- name: gemma7b-gpt4o_1k_summarize-lora
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# gemma7b-gpt4o_1k_summarize-lora
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 2.3456
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 0.0003
|
43 |
+
- train_batch_size: 1
|
44 |
+
- eval_batch_size: 2
|
45 |
+
- seed: 42
|
46 |
+
- distributed_type: multi-GPU
|
47 |
+
- num_devices: 4
|
48 |
+
- gradient_accumulation_steps: 2
|
49 |
+
- total_train_batch_size: 8
|
50 |
+
- total_eval_batch_size: 8
|
51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
52 |
+
- lr_scheduler_type: cosine
|
53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
54 |
+
- num_epochs: 1
|
55 |
+
|
56 |
+
### Training results
|
57 |
+
|
58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
60 |
+
| 0.8301 | 0.9997 | 1567 | 2.3456 |
|
61 |
+
|
62 |
+
|
63 |
+
### Framework versions
|
64 |
+
|
65 |
+
- PEFT 0.12.0
|
66 |
+
- Transformers 4.44.2
|
67 |
+
- Pytorch 2.4.1+cu121
|
68 |
+
- Datasets 3.0.0
|
69 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.9996810207336523,
|
3 |
+
"total_flos": 6.008617478547046e+17,
|
4 |
+
"train_loss": 1.5171029353948178,
|
5 |
+
"train_runtime": 9750.6774,
|
6 |
+
"train_samples": 115376,
|
7 |
+
"train_samples_per_second": 1.286,
|
8 |
+
"train_steps_per_second": 0.161
|
9 |
+
}
|
runs/Sep23_22-52-56_2e80285b3fe9/events.out.tfevents.1727132349.2e80285b3fe9.120709.0
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d242847b7e71f3f0576c553597c236d25f387ff0414238dc854b004c051152a7
|
3 |
+
size 73415
|
train_results.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"epoch": 0.9996810207336523,
|
3 |
+
"total_flos": 6.008617478547046e+17,
|
4 |
+
"train_loss": 1.5171029353948178,
|
5 |
+
"train_runtime": 9750.6774,
|
6 |
+
"train_samples": 115376,
|
7 |
+
"train_samples_per_second": 1.286,
|
8 |
+
"train_steps_per_second": 0.161
|
9 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,2248 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.9996810207336523,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1567,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.0006379585326953748,
|
13 |
+
"grad_norm": 694.1307983398438,
|
14 |
+
"learning_rate": 1.910828025477707e-06,
|
15 |
+
"loss": 45.2944,
|
16 |
+
"step": 1
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.003189792663476874,
|
20 |
+
"grad_norm": 503.5704345703125,
|
21 |
+
"learning_rate": 9.554140127388534e-06,
|
22 |
+
"loss": 45.6151,
|
23 |
+
"step": 5
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.006379585326953748,
|
27 |
+
"grad_norm": 180.77162170410156,
|
28 |
+
"learning_rate": 1.9108280254777068e-05,
|
29 |
+
"loss": 33.1885,
|
30 |
+
"step": 10
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.009569377990430622,
|
34 |
+
"grad_norm": 67.5441665649414,
|
35 |
+
"learning_rate": 2.8662420382165606e-05,
|
36 |
+
"loss": 23.199,
|
37 |
+
"step": 15
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.012759170653907496,
|
41 |
+
"grad_norm": 18.643754959106445,
|
42 |
+
"learning_rate": 3.8216560509554137e-05,
|
43 |
+
"loss": 20.7118,
|
44 |
+
"step": 20
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.01594896331738437,
|
48 |
+
"grad_norm": 10.430642127990723,
|
49 |
+
"learning_rate": 4.777070063694267e-05,
|
50 |
+
"loss": 19.5913,
|
51 |
+
"step": 25
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.019138755980861243,
|
55 |
+
"grad_norm": 13.150680541992188,
|
56 |
+
"learning_rate": 5.732484076433121e-05,
|
57 |
+
"loss": 18.0748,
|
58 |
+
"step": 30
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.022328548644338118,
|
62 |
+
"grad_norm": 34.60968017578125,
|
63 |
+
"learning_rate": 6.687898089171974e-05,
|
64 |
+
"loss": 15.8287,
|
65 |
+
"step": 35
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.025518341307814992,
|
69 |
+
"grad_norm": 77.59700775146484,
|
70 |
+
"learning_rate": 7.643312101910827e-05,
|
71 |
+
"loss": 11.4445,
|
72 |
+
"step": 40
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.028708133971291867,
|
76 |
+
"grad_norm": 14.57149600982666,
|
77 |
+
"learning_rate": 8.59872611464968e-05,
|
78 |
+
"loss": 3.2777,
|
79 |
+
"step": 45
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.03189792663476874,
|
83 |
+
"grad_norm": 6.958239555358887,
|
84 |
+
"learning_rate": 9.554140127388533e-05,
|
85 |
+
"loss": 2.233,
|
86 |
+
"step": 50
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.03508771929824561,
|
90 |
+
"grad_norm": 109.23677062988281,
|
91 |
+
"learning_rate": 0.00010509554140127387,
|
92 |
+
"loss": 1.8566,
|
93 |
+
"step": 55
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.03827751196172249,
|
97 |
+
"grad_norm": 12.268388748168945,
|
98 |
+
"learning_rate": 0.00011464968152866242,
|
99 |
+
"loss": 1.7099,
|
100 |
+
"step": 60
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.04146730462519936,
|
104 |
+
"grad_norm": 3.5562500953674316,
|
105 |
+
"learning_rate": 0.00012420382165605095,
|
106 |
+
"loss": 1.5828,
|
107 |
+
"step": 65
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.044657097288676235,
|
111 |
+
"grad_norm": 2.0333609580993652,
|
112 |
+
"learning_rate": 0.00013375796178343948,
|
113 |
+
"loss": 1.4119,
|
114 |
+
"step": 70
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.04784688995215311,
|
118 |
+
"grad_norm": 4.240214824676514,
|
119 |
+
"learning_rate": 0.00014331210191082802,
|
120 |
+
"loss": 1.3318,
|
121 |
+
"step": 75
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.051036682615629984,
|
125 |
+
"grad_norm": 5.074183940887451,
|
126 |
+
"learning_rate": 0.00015286624203821655,
|
127 |
+
"loss": 1.2783,
|
128 |
+
"step": 80
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.05422647527910686,
|
132 |
+
"grad_norm": 27.86067771911621,
|
133 |
+
"learning_rate": 0.00016242038216560508,
|
134 |
+
"loss": 1.2497,
|
135 |
+
"step": 85
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.05741626794258373,
|
139 |
+
"grad_norm": 4.769386291503906,
|
140 |
+
"learning_rate": 0.0001719745222929936,
|
141 |
+
"loss": 1.2015,
|
142 |
+
"step": 90
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.06060606060606061,
|
146 |
+
"grad_norm": 8.017805099487305,
|
147 |
+
"learning_rate": 0.00018152866242038214,
|
148 |
+
"loss": 1.1433,
|
149 |
+
"step": 95
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.06379585326953748,
|
153 |
+
"grad_norm": 10.532003402709961,
|
154 |
+
"learning_rate": 0.00019108280254777067,
|
155 |
+
"loss": 1.1267,
|
156 |
+
"step": 100
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.06698564593301436,
|
160 |
+
"grad_norm": 13.3386869430542,
|
161 |
+
"learning_rate": 0.0002006369426751592,
|
162 |
+
"loss": 1.1991,
|
163 |
+
"step": 105
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.07017543859649122,
|
167 |
+
"grad_norm": 6.093812942504883,
|
168 |
+
"learning_rate": 0.00021019108280254773,
|
169 |
+
"loss": 1.1734,
|
170 |
+
"step": 110
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.0733652312599681,
|
174 |
+
"grad_norm": 5.6636199951171875,
|
175 |
+
"learning_rate": 0.00021974522292993626,
|
176 |
+
"loss": 1.1269,
|
177 |
+
"step": 115
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.07655502392344497,
|
181 |
+
"grad_norm": 3.0813510417938232,
|
182 |
+
"learning_rate": 0.00022929936305732485,
|
183 |
+
"loss": 1.1113,
|
184 |
+
"step": 120
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.07974481658692185,
|
188 |
+
"grad_norm": 2.015939474105835,
|
189 |
+
"learning_rate": 0.00023885350318471338,
|
190 |
+
"loss": 1.0849,
|
191 |
+
"step": 125
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.08293460925039872,
|
195 |
+
"grad_norm": 9.842239379882812,
|
196 |
+
"learning_rate": 0.0002484076433121019,
|
197 |
+
"loss": 1.1141,
|
198 |
+
"step": 130
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.0861244019138756,
|
202 |
+
"grad_norm": 4.175192356109619,
|
203 |
+
"learning_rate": 0.00025796178343949044,
|
204 |
+
"loss": 1.0866,
|
205 |
+
"step": 135
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.08931419457735247,
|
209 |
+
"grad_norm": 7.420131206512451,
|
210 |
+
"learning_rate": 0.00026751592356687897,
|
211 |
+
"loss": 1.053,
|
212 |
+
"step": 140
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.09250398724082935,
|
216 |
+
"grad_norm": 20.004467010498047,
|
217 |
+
"learning_rate": 0.0002770700636942675,
|
218 |
+
"loss": 1.0895,
|
219 |
+
"step": 145
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.09569377990430622,
|
223 |
+
"grad_norm": 16.957359313964844,
|
224 |
+
"learning_rate": 0.00028662420382165603,
|
225 |
+
"loss": 1.0669,
|
226 |
+
"step": 150
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.09888357256778309,
|
230 |
+
"grad_norm": 3.6915156841278076,
|
231 |
+
"learning_rate": 0.00029617834394904456,
|
232 |
+
"loss": 1.0751,
|
233 |
+
"step": 155
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.10207336523125997,
|
237 |
+
"grad_norm": 4.615033149719238,
|
238 |
+
"learning_rate": 0.0002999966490829603,
|
239 |
+
"loss": 1.0461,
|
240 |
+
"step": 160
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.10526315789473684,
|
244 |
+
"grad_norm": 2.876262903213501,
|
245 |
+
"learning_rate": 0.00029997617179878324,
|
246 |
+
"loss": 1.0378,
|
247 |
+
"step": 165
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 0.10845295055821372,
|
251 |
+
"grad_norm": 3.2917582988739014,
|
252 |
+
"learning_rate": 0.0002999370813893047,
|
253 |
+
"loss": 1.0302,
|
254 |
+
"step": 170
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.11164274322169059,
|
258 |
+
"grad_norm": 7.9203715324401855,
|
259 |
+
"learning_rate": 0.00029987938270592676,
|
260 |
+
"loss": 1.067,
|
261 |
+
"step": 175
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 0.11483253588516747,
|
265 |
+
"grad_norm": 4.65939474105835,
|
266 |
+
"learning_rate": 0.0002998030829094724,
|
267 |
+
"loss": 1.0646,
|
268 |
+
"step": 180
|
269 |
+
},
|
270 |
+
{
|
271 |
+
"epoch": 0.11802232854864433,
|
272 |
+
"grad_norm": 1.8894805908203125,
|
273 |
+
"learning_rate": 0.00029970819146929694,
|
274 |
+
"loss": 1.0093,
|
275 |
+
"step": 185
|
276 |
+
},
|
277 |
+
{
|
278 |
+
"epoch": 0.12121212121212122,
|
279 |
+
"grad_norm": 2.7538986206054688,
|
280 |
+
"learning_rate": 0.0002995947201621131,
|
281 |
+
"loss": 1.0402,
|
282 |
+
"step": 190
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 0.12440191387559808,
|
286 |
+
"grad_norm": 2.2546260356903076,
|
287 |
+
"learning_rate": 0.00029946268307052886,
|
288 |
+
"loss": 1.0247,
|
289 |
+
"step": 195
|
290 |
+
},
|
291 |
+
{
|
292 |
+
"epoch": 0.12759170653907495,
|
293 |
+
"grad_norm": 2.655927896499634,
|
294 |
+
"learning_rate": 0.0002993120965813003,
|
295 |
+
"loss": 0.9994,
|
296 |
+
"step": 200
|
297 |
+
},
|
298 |
+
{
|
299 |
+
"epoch": 0.13078149920255183,
|
300 |
+
"grad_norm": 1.8121349811553955,
|
301 |
+
"learning_rate": 0.0002991429793832975,
|
302 |
+
"loss": 1.0279,
|
303 |
+
"step": 205
|
304 |
+
},
|
305 |
+
{
|
306 |
+
"epoch": 0.1339712918660287,
|
307 |
+
"grad_norm": 1.38643217086792,
|
308 |
+
"learning_rate": 0.00029895535246518524,
|
309 |
+
"loss": 1.0534,
|
310 |
+
"step": 210
|
311 |
+
},
|
312 |
+
{
|
313 |
+
"epoch": 0.1371610845295056,
|
314 |
+
"grad_norm": 4.24297571182251,
|
315 |
+
"learning_rate": 0.0002987492391128182,
|
316 |
+
"loss": 1.0622,
|
317 |
+
"step": 215
|
318 |
+
},
|
319 |
+
{
|
320 |
+
"epoch": 0.14035087719298245,
|
321 |
+
"grad_norm": 2.3828201293945312,
|
322 |
+
"learning_rate": 0.0002985246649063509,
|
323 |
+
"loss": 1.0239,
|
324 |
+
"step": 220
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 0.14354066985645933,
|
328 |
+
"grad_norm": 1.4523991346359253,
|
329 |
+
"learning_rate": 0.0002982816577170631,
|
330 |
+
"loss": 1.026,
|
331 |
+
"step": 225
|
332 |
+
},
|
333 |
+
{
|
334 |
+
"epoch": 0.1467304625199362,
|
335 |
+
"grad_norm": 1.3892016410827637,
|
336 |
+
"learning_rate": 0.00029802024770390087,
|
337 |
+
"loss": 1.0298,
|
338 |
+
"step": 230
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.14992025518341306,
|
342 |
+
"grad_norm": 3.4726333618164062,
|
343 |
+
"learning_rate": 0.00029774046730973334,
|
344 |
+
"loss": 1.0316,
|
345 |
+
"step": 235
|
346 |
+
},
|
347 |
+
{
|
348 |
+
"epoch": 0.15311004784688995,
|
349 |
+
"grad_norm": 2.9531638622283936,
|
350 |
+
"learning_rate": 0.00029744235125732664,
|
351 |
+
"loss": 1.0196,
|
352 |
+
"step": 240
|
353 |
+
},
|
354 |
+
{
|
355 |
+
"epoch": 0.15629984051036683,
|
356 |
+
"grad_norm": 1.494456171989441,
|
357 |
+
"learning_rate": 0.0002971259365450344,
|
358 |
+
"loss": 1.0245,
|
359 |
+
"step": 245
|
360 |
+
},
|
361 |
+
{
|
362 |
+
"epoch": 0.1594896331738437,
|
363 |
+
"grad_norm": 3.462984323501587,
|
364 |
+
"learning_rate": 0.00029679126244220596,
|
365 |
+
"loss": 0.9657,
|
366 |
+
"step": 250
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 0.16267942583732056,
|
370 |
+
"grad_norm": 2.6760306358337402,
|
371 |
+
"learning_rate": 0.00029643837048431293,
|
372 |
+
"loss": 0.9942,
|
373 |
+
"step": 255
|
374 |
+
},
|
375 |
+
{
|
376 |
+
"epoch": 0.16586921850079744,
|
377 |
+
"grad_norm": 3.4000234603881836,
|
378 |
+
"learning_rate": 0.0002960673044677939,
|
379 |
+
"loss": 0.9955,
|
380 |
+
"step": 260
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.16905901116427433,
|
384 |
+
"grad_norm": 2.6364219188690186,
|
385 |
+
"learning_rate": 0.00029567811044461977,
|
386 |
+
"loss": 0.966,
|
387 |
+
"step": 265
|
388 |
+
},
|
389 |
+
{
|
390 |
+
"epoch": 0.1722488038277512,
|
391 |
+
"grad_norm": 10.39656925201416,
|
392 |
+
"learning_rate": 0.00029527083671657746,
|
393 |
+
"loss": 0.9989,
|
394 |
+
"step": 270
|
395 |
+
},
|
396 |
+
{
|
397 |
+
"epoch": 0.17543859649122806,
|
398 |
+
"grad_norm": 2.5434165000915527,
|
399 |
+
"learning_rate": 0.00029484553382927594,
|
400 |
+
"loss": 0.9809,
|
401 |
+
"step": 275
|
402 |
+
},
|
403 |
+
{
|
404 |
+
"epoch": 0.17862838915470494,
|
405 |
+
"grad_norm": 1.5303857326507568,
|
406 |
+
"learning_rate": 0.000294402254565873,
|
407 |
+
"loss": 1.0049,
|
408 |
+
"step": 280
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 0.18181818181818182,
|
412 |
+
"grad_norm": 5.918676376342773,
|
413 |
+
"learning_rate": 0.00029394105394052434,
|
414 |
+
"loss": 0.995,
|
415 |
+
"step": 285
|
416 |
+
},
|
417 |
+
{
|
418 |
+
"epoch": 0.1850079744816587,
|
419 |
+
"grad_norm": 1.4962425231933594,
|
420 |
+
"learning_rate": 0.00029346198919155616,
|
421 |
+
"loss": 1.0264,
|
422 |
+
"step": 290
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.18819776714513556,
|
426 |
+
"grad_norm": 1.3401553630828857,
|
427 |
+
"learning_rate": 0.00029296511977436107,
|
428 |
+
"loss": 0.998,
|
429 |
+
"step": 295
|
430 |
+
},
|
431 |
+
{
|
432 |
+
"epoch": 0.19138755980861244,
|
433 |
+
"grad_norm": 3.1120340824127197,
|
434 |
+
"learning_rate": 0.0002924505073540198,
|
435 |
+
"loss": 0.9931,
|
436 |
+
"step": 300
|
437 |
+
},
|
438 |
+
{
|
439 |
+
"epoch": 0.19457735247208932,
|
440 |
+
"grad_norm": 2.893742561340332,
|
441 |
+
"learning_rate": 0.0002919182157976476,
|
442 |
+
"loss": 0.9924,
|
443 |
+
"step": 305
|
444 |
+
},
|
445 |
+
{
|
446 |
+
"epoch": 0.19776714513556617,
|
447 |
+
"grad_norm": 3.6078243255615234,
|
448 |
+
"learning_rate": 0.00029136831116646815,
|
449 |
+
"loss": 0.9973,
|
450 |
+
"step": 310
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 0.20095693779904306,
|
454 |
+
"grad_norm": 1.5071393251419067,
|
455 |
+
"learning_rate": 0.000290800861707615,
|
456 |
+
"loss": 0.9707,
|
457 |
+
"step": 315
|
458 |
+
},
|
459 |
+
{
|
460 |
+
"epoch": 0.20414673046251994,
|
461 |
+
"grad_norm": 20.275678634643555,
|
462 |
+
"learning_rate": 0.00029021593784566113,
|
463 |
+
"loss": 0.9897,
|
464 |
+
"step": 320
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.20733652312599682,
|
468 |
+
"grad_norm": 1.4310753345489502,
|
469 |
+
"learning_rate": 0.0002896136121738793,
|
470 |
+
"loss": 0.9855,
|
471 |
+
"step": 325
|
472 |
+
},
|
473 |
+
{
|
474 |
+
"epoch": 0.21052631578947367,
|
475 |
+
"grad_norm": 2.6854515075683594,
|
476 |
+
"learning_rate": 0.0002889939594452323,
|
477 |
+
"loss": 0.9701,
|
478 |
+
"step": 330
|
479 |
+
},
|
480 |
+
{
|
481 |
+
"epoch": 0.21371610845295055,
|
482 |
+
"grad_norm": 1.3763470649719238,
|
483 |
+
"learning_rate": 0.00028835705656309583,
|
484 |
+
"loss": 0.9812,
|
485 |
+
"step": 335
|
486 |
+
},
|
487 |
+
{
|
488 |
+
"epoch": 0.21690590111642744,
|
489 |
+
"grad_norm": 1.8561792373657227,
|
490 |
+
"learning_rate": 0.0002877029825717142,
|
491 |
+
"loss": 0.9943,
|
492 |
+
"step": 340
|
493 |
+
},
|
494 |
+
{
|
495 |
+
"epoch": 0.22009569377990432,
|
496 |
+
"grad_norm": 3.642995595932007,
|
497 |
+
"learning_rate": 0.0002870318186463901,
|
498 |
+
"loss": 0.9637,
|
499 |
+
"step": 345
|
500 |
+
},
|
501 |
+
{
|
502 |
+
"epoch": 0.22328548644338117,
|
503 |
+
"grad_norm": 4.639859199523926,
|
504 |
+
"learning_rate": 0.0002863436480834105,
|
505 |
+
"loss": 0.9998,
|
506 |
+
"step": 350
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.22647527910685805,
|
510 |
+
"grad_norm": 1.1026989221572876,
|
511 |
+
"learning_rate": 0.00028563855628970886,
|
512 |
+
"loss": 0.9777,
|
513 |
+
"step": 355
|
514 |
+
},
|
515 |
+
{
|
516 |
+
"epoch": 0.22966507177033493,
|
517 |
+
"grad_norm": 1.1374722719192505,
|
518 |
+
"learning_rate": 0.0002849166307722653,
|
519 |
+
"loss": 1.0051,
|
520 |
+
"step": 360
|
521 |
+
},
|
522 |
+
{
|
523 |
+
"epoch": 0.23285486443381181,
|
524 |
+
"grad_norm": 1.255344033241272,
|
525 |
+
"learning_rate": 0.00028417796112724684,
|
526 |
+
"loss": 1.009,
|
527 |
+
"step": 365
|
528 |
+
},
|
529 |
+
{
|
530 |
+
"epoch": 0.23604465709728867,
|
531 |
+
"grad_norm": 1.4702554941177368,
|
532 |
+
"learning_rate": 0.0002834226390288873,
|
533 |
+
"loss": 0.9616,
|
534 |
+
"step": 370
|
535 |
+
},
|
536 |
+
{
|
537 |
+
"epoch": 0.23923444976076555,
|
538 |
+
"grad_norm": 1.4822920560836792,
|
539 |
+
"learning_rate": 0.0002826507582181103,
|
540 |
+
"loss": 0.966,
|
541 |
+
"step": 375
|
542 |
+
},
|
543 |
+
{
|
544 |
+
"epoch": 0.24242424242424243,
|
545 |
+
"grad_norm": 6.145591735839844,
|
546 |
+
"learning_rate": 0.00028186241449089524,
|
547 |
+
"loss": 0.9819,
|
548 |
+
"step": 380
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.24561403508771928,
|
552 |
+
"grad_norm": 1.982974886894226,
|
553 |
+
"learning_rate": 0.000281057705686388,
|
554 |
+
"loss": 0.9766,
|
555 |
+
"step": 385
|
556 |
+
},
|
557 |
+
{
|
558 |
+
"epoch": 0.24880382775119617,
|
559 |
+
"grad_norm": 1.2046759128570557,
|
560 |
+
"learning_rate": 0.0002802367316747589,
|
561 |
+
"loss": 0.9704,
|
562 |
+
"step": 390
|
563 |
+
},
|
564 |
+
{
|
565 |
+
"epoch": 0.25199362041467305,
|
566 |
+
"grad_norm": 1.5054329633712769,
|
567 |
+
"learning_rate": 0.0002793995943448078,
|
568 |
+
"loss": 0.9484,
|
569 |
+
"step": 395
|
570 |
+
},
|
571 |
+
{
|
572 |
+
"epoch": 0.2551834130781499,
|
573 |
+
"grad_norm": 7.931117534637451,
|
574 |
+
"learning_rate": 0.00027854639759131893,
|
575 |
+
"loss": 0.9548,
|
576 |
+
"step": 400
|
577 |
+
},
|
578 |
+
{
|
579 |
+
"epoch": 0.2583732057416268,
|
580 |
+
"grad_norm": 1.8060181140899658,
|
581 |
+
"learning_rate": 0.00027767724730216696,
|
582 |
+
"loss": 0.9946,
|
583 |
+
"step": 405
|
584 |
+
},
|
585 |
+
{
|
586 |
+
"epoch": 0.26156299840510366,
|
587 |
+
"grad_norm": 2.3200082778930664,
|
588 |
+
"learning_rate": 0.0002767922513451754,
|
589 |
+
"loss": 0.9888,
|
590 |
+
"step": 410
|
591 |
+
},
|
592 |
+
{
|
593 |
+
"epoch": 0.2647527910685805,
|
594 |
+
"grad_norm": 2.2360453605651855,
|
595 |
+
"learning_rate": 0.00027589151955472965,
|
596 |
+
"loss": 0.9588,
|
597 |
+
"step": 415
|
598 |
+
},
|
599 |
+
{
|
600 |
+
"epoch": 0.2679425837320574,
|
601 |
+
"grad_norm": 1.594473123550415,
|
602 |
+
"learning_rate": 0.00027497516371814543,
|
603 |
+
"loss": 0.9547,
|
604 |
+
"step": 420
|
605 |
+
},
|
606 |
+
{
|
607 |
+
"epoch": 0.2711323763955343,
|
608 |
+
"grad_norm": 3.532134532928467,
|
609 |
+
"learning_rate": 0.00027404329756179537,
|
610 |
+
"loss": 0.9589,
|
611 |
+
"step": 425
|
612 |
+
},
|
613 |
+
{
|
614 |
+
"epoch": 0.2743221690590112,
|
615 |
+
"grad_norm": 2.1840810775756836,
|
616 |
+
"learning_rate": 0.0002730960367369949,
|
617 |
+
"loss": 1.017,
|
618 |
+
"step": 430
|
619 |
+
},
|
620 |
+
{
|
621 |
+
"epoch": 0.27751196172248804,
|
622 |
+
"grad_norm": 1.9814718961715698,
|
623 |
+
"learning_rate": 0.00027213349880564873,
|
624 |
+
"loss": 0.9901,
|
625 |
+
"step": 435
|
626 |
+
},
|
627 |
+
{
|
628 |
+
"epoch": 0.2807017543859649,
|
629 |
+
"grad_norm": 2.3927841186523438,
|
630 |
+
"learning_rate": 0.0002711558032256607,
|
631 |
+
"loss": 0.9715,
|
632 |
+
"step": 440
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.2838915470494418,
|
636 |
+
"grad_norm": 1.501244306564331,
|
637 |
+
"learning_rate": 0.0002701630713361085,
|
638 |
+
"loss": 0.9621,
|
639 |
+
"step": 445
|
640 |
+
},
|
641 |
+
{
|
642 |
+
"epoch": 0.28708133971291866,
|
643 |
+
"grad_norm": 1.6648037433624268,
|
644 |
+
"learning_rate": 0.00026915542634218403,
|
645 |
+
"loss": 0.9676,
|
646 |
+
"step": 450
|
647 |
+
},
|
648 |
+
{
|
649 |
+
"epoch": 0.2902711323763955,
|
650 |
+
"grad_norm": 2.717752695083618,
|
651 |
+
"learning_rate": 0.00026813299329990335,
|
652 |
+
"loss": 0.9837,
|
653 |
+
"step": 455
|
654 |
+
},
|
655 |
+
{
|
656 |
+
"epoch": 0.2934609250398724,
|
657 |
+
"grad_norm": 1.7805887460708618,
|
658 |
+
"learning_rate": 0.0002670958991005859,
|
659 |
+
"loss": 0.9635,
|
660 |
+
"step": 460
|
661 |
+
},
|
662 |
+
{
|
663 |
+
"epoch": 0.2966507177033493,
|
664 |
+
"grad_norm": 2.2134859561920166,
|
665 |
+
"learning_rate": 0.0002660442724551065,
|
666 |
+
"loss": 0.9566,
|
667 |
+
"step": 465
|
668 |
+
},
|
669 |
+
{
|
670 |
+
"epoch": 0.29984051036682613,
|
671 |
+
"grad_norm": 1.2682667970657349,
|
672 |
+
"learning_rate": 0.00026497824387792146,
|
673 |
+
"loss": 0.963,
|
674 |
+
"step": 470
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.30303030303030304,
|
678 |
+
"grad_norm": 1.2729476690292358,
|
679 |
+
"learning_rate": 0.00026389794567087085,
|
680 |
+
"loss": 0.964,
|
681 |
+
"step": 475
|
682 |
+
},
|
683 |
+
{
|
684 |
+
"epoch": 0.3062200956937799,
|
685 |
+
"grad_norm": 1.4577951431274414,
|
686 |
+
"learning_rate": 0.0002628035119067586,
|
687 |
+
"loss": 0.9487,
|
688 |
+
"step": 480
|
689 |
+
},
|
690 |
+
{
|
691 |
+
"epoch": 0.3094098883572568,
|
692 |
+
"grad_norm": 1.637868881225586,
|
693 |
+
"learning_rate": 0.0002616950784127135,
|
694 |
+
"loss": 0.9453,
|
695 |
+
"step": 485
|
696 |
+
},
|
697 |
+
{
|
698 |
+
"epoch": 0.31259968102073366,
|
699 |
+
"grad_norm": 1.3225361108779907,
|
700 |
+
"learning_rate": 0.00026057278275333165,
|
701 |
+
"loss": 0.9814,
|
702 |
+
"step": 490
|
703 |
+
},
|
704 |
+
{
|
705 |
+
"epoch": 0.3157894736842105,
|
706 |
+
"grad_norm": 1.1870254278182983,
|
707 |
+
"learning_rate": 0.00025943676421360395,
|
708 |
+
"loss": 0.9555,
|
709 |
+
"step": 495
|
710 |
+
},
|
711 |
+
{
|
712 |
+
"epoch": 0.3189792663476874,
|
713 |
+
"grad_norm": 2.6951119899749756,
|
714 |
+
"learning_rate": 0.00025828716378163,
|
715 |
+
"loss": 0.9657,
|
716 |
+
"step": 500
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.32216905901116427,
|
720 |
+
"grad_norm": 2.149862766265869,
|
721 |
+
"learning_rate": 0.00025712412413112006,
|
722 |
+
"loss": 0.9697,
|
723 |
+
"step": 505
|
724 |
+
},
|
725 |
+
{
|
726 |
+
"epoch": 0.3253588516746411,
|
727 |
+
"grad_norm": 1.2473626136779785,
|
728 |
+
"learning_rate": 0.00025594778960368844,
|
729 |
+
"loss": 0.9743,
|
730 |
+
"step": 510
|
731 |
+
},
|
732 |
+
{
|
733 |
+
"epoch": 0.32854864433811803,
|
734 |
+
"grad_norm": 1.7699605226516724,
|
735 |
+
"learning_rate": 0.0002547583061909396,
|
736 |
+
"loss": 0.9564,
|
737 |
+
"step": 515
|
738 |
+
},
|
739 |
+
{
|
740 |
+
"epoch": 0.3317384370015949,
|
741 |
+
"grad_norm": 2.645415782928467,
|
742 |
+
"learning_rate": 0.00025355582151634956,
|
743 |
+
"loss": 0.9812,
|
744 |
+
"step": 520
|
745 |
+
},
|
746 |
+
{
|
747 |
+
"epoch": 0.3349282296650718,
|
748 |
+
"grad_norm": 1.1734089851379395,
|
749 |
+
"learning_rate": 0.00025234048481694477,
|
750 |
+
"loss": 0.9463,
|
751 |
+
"step": 525
|
752 |
+
},
|
753 |
+
{
|
754 |
+
"epoch": 0.33811802232854865,
|
755 |
+
"grad_norm": 1.7658518552780151,
|
756 |
+
"learning_rate": 0.0002511124469247809,
|
757 |
+
"loss": 0.9547,
|
758 |
+
"step": 530
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.3413078149920255,
|
762 |
+
"grad_norm": 1.289113163948059,
|
763 |
+
"learning_rate": 0.00024987186024822295,
|
764 |
+
"loss": 0.9426,
|
765 |
+
"step": 535
|
766 |
+
},
|
767 |
+
{
|
768 |
+
"epoch": 0.3444976076555024,
|
769 |
+
"grad_norm": 10.93489933013916,
|
770 |
+
"learning_rate": 0.0002486188787530309,
|
771 |
+
"loss": 0.9476,
|
772 |
+
"step": 540
|
773 |
+
},
|
774 |
+
{
|
775 |
+
"epoch": 0.34768740031897927,
|
776 |
+
"grad_norm": 1.6490458250045776,
|
777 |
+
"learning_rate": 0.00024735365794325117,
|
778 |
+
"loss": 0.976,
|
779 |
+
"step": 545
|
780 |
+
},
|
781 |
+
{
|
782 |
+
"epoch": 0.3508771929824561,
|
783 |
+
"grad_norm": 1.1837722063064575,
|
784 |
+
"learning_rate": 0.0002460763548419172,
|
785 |
+
"loss": 0.9391,
|
786 |
+
"step": 550
|
787 |
+
},
|
788 |
+
{
|
789 |
+
"epoch": 0.35406698564593303,
|
790 |
+
"grad_norm": 1.1305694580078125,
|
791 |
+
"learning_rate": 0.0002447871279715624,
|
792 |
+
"loss": 0.9555,
|
793 |
+
"step": 555
|
794 |
+
},
|
795 |
+
{
|
796 |
+
"epoch": 0.3572567783094099,
|
797 |
+
"grad_norm": 1.6349717378616333,
|
798 |
+
"learning_rate": 0.00024348613733454565,
|
799 |
+
"loss": 0.9412,
|
800 |
+
"step": 560
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 0.36044657097288674,
|
804 |
+
"grad_norm": 2.098814010620117,
|
805 |
+
"learning_rate": 0.00024217354439319427,
|
806 |
+
"loss": 0.9365,
|
807 |
+
"step": 565
|
808 |
+
},
|
809 |
+
{
|
810 |
+
"epoch": 0.36363636363636365,
|
811 |
+
"grad_norm": 1.4904630184173584,
|
812 |
+
"learning_rate": 0.00024084951204976528,
|
813 |
+
"loss": 0.9243,
|
814 |
+
"step": 570
|
815 |
+
},
|
816 |
+
{
|
817 |
+
"epoch": 0.3668261562998405,
|
818 |
+
"grad_norm": 2.3611807823181152,
|
819 |
+
"learning_rate": 0.0002395142046262281,
|
820 |
+
"loss": 0.9613,
|
821 |
+
"step": 575
|
822 |
+
},
|
823 |
+
{
|
824 |
+
"epoch": 0.3700159489633174,
|
825 |
+
"grad_norm": 2.1644859313964844,
|
826 |
+
"learning_rate": 0.00023816778784387094,
|
827 |
+
"loss": 0.942,
|
828 |
+
"step": 580
|
829 |
+
},
|
830 |
+
{
|
831 |
+
"epoch": 0.37320574162679426,
|
832 |
+
"grad_norm": 1.9585081338882446,
|
833 |
+
"learning_rate": 0.0002368104288027336,
|
834 |
+
"loss": 0.9402,
|
835 |
+
"step": 585
|
836 |
+
},
|
837 |
+
{
|
838 |
+
"epoch": 0.3763955342902711,
|
839 |
+
"grad_norm": 1.793800950050354,
|
840 |
+
"learning_rate": 0.0002354422959608692,
|
841 |
+
"loss": 0.9421,
|
842 |
+
"step": 590
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 0.379585326953748,
|
846 |
+
"grad_norm": 2.8529624938964844,
|
847 |
+
"learning_rate": 0.00023406355911343717,
|
848 |
+
"loss": 0.9688,
|
849 |
+
"step": 595
|
850 |
+
},
|
851 |
+
{
|
852 |
+
"epoch": 0.3827751196172249,
|
853 |
+
"grad_norm": 1.6029831171035767,
|
854 |
+
"learning_rate": 0.00023267438937163077,
|
855 |
+
"loss": 0.9164,
|
856 |
+
"step": 600
|
857 |
+
},
|
858 |
+
{
|
859 |
+
"epoch": 0.38596491228070173,
|
860 |
+
"grad_norm": 1.7395401000976562,
|
861 |
+
"learning_rate": 0.00023127495914144051,
|
862 |
+
"loss": 0.9409,
|
863 |
+
"step": 605
|
864 |
+
},
|
865 |
+
{
|
866 |
+
"epoch": 0.38915470494417864,
|
867 |
+
"grad_norm": 1.7709709405899048,
|
868 |
+
"learning_rate": 0.00022986544210225774,
|
869 |
+
"loss": 0.9353,
|
870 |
+
"step": 610
|
871 |
+
},
|
872 |
+
{
|
873 |
+
"epoch": 0.3923444976076555,
|
874 |
+
"grad_norm": 1.1406283378601074,
|
875 |
+
"learning_rate": 0.00022844601318531955,
|
876 |
+
"loss": 0.9163,
|
877 |
+
"step": 615
|
878 |
+
},
|
879 |
+
{
|
880 |
+
"epoch": 0.39553429027113235,
|
881 |
+
"grad_norm": 1.617177963256836,
|
882 |
+
"learning_rate": 0.00022701684855199857,
|
883 |
+
"loss": 0.961,
|
884 |
+
"step": 620
|
885 |
+
},
|
886 |
+
{
|
887 |
+
"epoch": 0.39872408293460926,
|
888 |
+
"grad_norm": 1.387049674987793,
|
889 |
+
"learning_rate": 0.00022557812557194,
|
890 |
+
"loss": 0.9273,
|
891 |
+
"step": 625
|
892 |
+
},
|
893 |
+
{
|
894 |
+
"epoch": 0.4019138755980861,
|
895 |
+
"grad_norm": 1.331063151359558,
|
896 |
+
"learning_rate": 0.00022413002280104915,
|
897 |
+
"loss": 0.9132,
|
898 |
+
"step": 630
|
899 |
+
},
|
900 |
+
{
|
901 |
+
"epoch": 0.405103668261563,
|
902 |
+
"grad_norm": 1.5018112659454346,
|
903 |
+
"learning_rate": 0.00022267271995933074,
|
904 |
+
"loss": 0.947,
|
905 |
+
"step": 635
|
906 |
+
},
|
907 |
+
{
|
908 |
+
"epoch": 0.4082934609250399,
|
909 |
+
"grad_norm": 1.384469747543335,
|
910 |
+
"learning_rate": 0.00022120639790858482,
|
911 |
+
"loss": 0.9328,
|
912 |
+
"step": 640
|
913 |
+
},
|
914 |
+
{
|
915 |
+
"epoch": 0.41148325358851673,
|
916 |
+
"grad_norm": 1.715080738067627,
|
917 |
+
"learning_rate": 0.00021973123862996044,
|
918 |
+
"loss": 0.9463,
|
919 |
+
"step": 645
|
920 |
+
},
|
921 |
+
{
|
922 |
+
"epoch": 0.41467304625199364,
|
923 |
+
"grad_norm": 1.2330074310302734,
|
924 |
+
"learning_rate": 0.00021824742520137026,
|
925 |
+
"loss": 0.9391,
|
926 |
+
"step": 650
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 0.4178628389154705,
|
930 |
+
"grad_norm": 1.498789668083191,
|
931 |
+
"learning_rate": 0.00021675514177476945,
|
932 |
+
"loss": 0.93,
|
933 |
+
"step": 655
|
934 |
+
},
|
935 |
+
{
|
936 |
+
"epoch": 0.42105263157894735,
|
937 |
+
"grad_norm": 1.287414789199829,
|
938 |
+
"learning_rate": 0.0002152545735533012,
|
939 |
+
"loss": 0.9554,
|
940 |
+
"step": 660
|
941 |
+
},
|
942 |
+
{
|
943 |
+
"epoch": 0.42424242424242425,
|
944 |
+
"grad_norm": 2.2374627590179443,
|
945 |
+
"learning_rate": 0.00021374590676831136,
|
946 |
+
"loss": 0.9394,
|
947 |
+
"step": 665
|
948 |
+
},
|
949 |
+
{
|
950 |
+
"epoch": 0.4274322169059011,
|
951 |
+
"grad_norm": 1.494278073310852,
|
952 |
+
"learning_rate": 0.00021222932865623605,
|
953 |
+
"loss": 0.9463,
|
954 |
+
"step": 670
|
955 |
+
},
|
956 |
+
{
|
957 |
+
"epoch": 0.430622009569378,
|
958 |
+
"grad_norm": 1.40542733669281,
|
959 |
+
"learning_rate": 0.00021070502743536414,
|
960 |
+
"loss": 0.9088,
|
961 |
+
"step": 675
|
962 |
+
},
|
963 |
+
{
|
964 |
+
"epoch": 0.43381180223285487,
|
965 |
+
"grad_norm": 1.9738633632659912,
|
966 |
+
"learning_rate": 0.00020917319228247805,
|
967 |
+
"loss": 0.9089,
|
968 |
+
"step": 680
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 0.4370015948963317,
|
972 |
+
"grad_norm": 2.0033769607543945,
|
973 |
+
"learning_rate": 0.00020763401330937555,
|
974 |
+
"loss": 0.9182,
|
975 |
+
"step": 685
|
976 |
+
},
|
977 |
+
{
|
978 |
+
"epoch": 0.44019138755980863,
|
979 |
+
"grad_norm": 1.3576220273971558,
|
980 |
+
"learning_rate": 0.00020608768153927546,
|
981 |
+
"loss": 0.9269,
|
982 |
+
"step": 690
|
983 |
+
},
|
984 |
+
{
|
985 |
+
"epoch": 0.4433811802232855,
|
986 |
+
"grad_norm": 1.2206989526748657,
|
987 |
+
"learning_rate": 0.00020453438888311042,
|
988 |
+
"loss": 0.9185,
|
989 |
+
"step": 695
|
990 |
+
},
|
991 |
+
{
|
992 |
+
"epoch": 0.44657097288676234,
|
993 |
+
"grad_norm": 1.5050541162490845,
|
994 |
+
"learning_rate": 0.00020297432811570916,
|
995 |
+
"loss": 0.9287,
|
996 |
+
"step": 700
|
997 |
+
},
|
998 |
+
{
|
999 |
+
"epoch": 0.44976076555023925,
|
1000 |
+
"grad_norm": 1.3409852981567383,
|
1001 |
+
"learning_rate": 0.00020140769285187187,
|
1002 |
+
"loss": 0.9089,
|
1003 |
+
"step": 705
|
1004 |
+
},
|
1005 |
+
{
|
1006 |
+
"epoch": 0.4529505582137161,
|
1007 |
+
"grad_norm": 3.004441738128662,
|
1008 |
+
"learning_rate": 0.00019983467752234132,
|
1009 |
+
"loss": 0.8978,
|
1010 |
+
"step": 710
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 0.45614035087719296,
|
1014 |
+
"grad_norm": 1.6164394617080688,
|
1015 |
+
"learning_rate": 0.0001982554773496723,
|
1016 |
+
"loss": 0.9464,
|
1017 |
+
"step": 715
|
1018 |
+
},
|
1019 |
+
{
|
1020 |
+
"epoch": 0.45933014354066987,
|
1021 |
+
"grad_norm": 1.1452794075012207,
|
1022 |
+
"learning_rate": 0.00019667028832400345,
|
1023 |
+
"loss": 0.8884,
|
1024 |
+
"step": 720
|
1025 |
+
},
|
1026 |
+
{
|
1027 |
+
"epoch": 0.4625199362041467,
|
1028 |
+
"grad_norm": 1.80891752243042,
|
1029 |
+
"learning_rate": 0.00019507930717873313,
|
1030 |
+
"loss": 0.9167,
|
1031 |
+
"step": 725
|
1032 |
+
},
|
1033 |
+
{
|
1034 |
+
"epoch": 0.46570972886762363,
|
1035 |
+
"grad_norm": 1.9527322053909302,
|
1036 |
+
"learning_rate": 0.00019348273136610364,
|
1037 |
+
"loss": 0.9182,
|
1038 |
+
"step": 730
|
1039 |
+
},
|
1040 |
+
{
|
1041 |
+
"epoch": 0.4688995215311005,
|
1042 |
+
"grad_norm": 1.8810184001922607,
|
1043 |
+
"learning_rate": 0.00019188075903269587,
|
1044 |
+
"loss": 0.9269,
|
1045 |
+
"step": 735
|
1046 |
+
},
|
1047 |
+
{
|
1048 |
+
"epoch": 0.47208931419457734,
|
1049 |
+
"grad_norm": 1.5535273551940918,
|
1050 |
+
"learning_rate": 0.00019027358899483776,
|
1051 |
+
"loss": 0.9048,
|
1052 |
+
"step": 740
|
1053 |
+
},
|
1054 |
+
{
|
1055 |
+
"epoch": 0.47527910685805425,
|
1056 |
+
"grad_norm": 1.7040493488311768,
|
1057 |
+
"learning_rate": 0.00018866142071393013,
|
1058 |
+
"loss": 0.9181,
|
1059 |
+
"step": 745
|
1060 |
+
},
|
1061 |
+
{
|
1062 |
+
"epoch": 0.4784688995215311,
|
1063 |
+
"grad_norm": 1.7220163345336914,
|
1064 |
+
"learning_rate": 0.00018704445427169156,
|
1065 |
+
"loss": 0.9242,
|
1066 |
+
"step": 750
|
1067 |
+
},
|
1068 |
+
{
|
1069 |
+
"epoch": 0.48165869218500795,
|
1070 |
+
"grad_norm": 1.5103449821472168,
|
1071 |
+
"learning_rate": 0.00018542289034532733,
|
1072 |
+
"loss": 0.9198,
|
1073 |
+
"step": 755
|
1074 |
+
},
|
1075 |
+
{
|
1076 |
+
"epoch": 0.48484848484848486,
|
1077 |
+
"grad_norm": 1.3625946044921875,
|
1078 |
+
"learning_rate": 0.00018379693018262349,
|
1079 |
+
"loss": 0.9253,
|
1080 |
+
"step": 760
|
1081 |
+
},
|
1082 |
+
{
|
1083 |
+
"epoch": 0.4880382775119617,
|
1084 |
+
"grad_norm": 1.1106137037277222,
|
1085 |
+
"learning_rate": 0.00018216677557697083,
|
1086 |
+
"loss": 0.9177,
|
1087 |
+
"step": 765
|
1088 |
+
},
|
1089 |
+
{
|
1090 |
+
"epoch": 0.49122807017543857,
|
1091 |
+
"grad_norm": 1.217457890510559,
|
1092 |
+
"learning_rate": 0.00018053262884232078,
|
1093 |
+
"loss": 0.9015,
|
1094 |
+
"step": 770
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 0.4944178628389155,
|
1098 |
+
"grad_norm": 1.1275889873504639,
|
1099 |
+
"learning_rate": 0.0001788946927880768,
|
1100 |
+
"loss": 0.9088,
|
1101 |
+
"step": 775
|
1102 |
+
},
|
1103 |
+
{
|
1104 |
+
"epoch": 0.49760765550239233,
|
1105 |
+
"grad_norm": 1.321915864944458,
|
1106 |
+
"learning_rate": 0.00017725317069392418,
|
1107 |
+
"loss": 0.8997,
|
1108 |
+
"step": 780
|
1109 |
+
},
|
1110 |
+
{
|
1111 |
+
"epoch": 0.5007974481658692,
|
1112 |
+
"grad_norm": 1.509143590927124,
|
1113 |
+
"learning_rate": 0.00017560826628460182,
|
1114 |
+
"loss": 0.9071,
|
1115 |
+
"step": 785
|
1116 |
+
},
|
1117 |
+
{
|
1118 |
+
"epoch": 0.5039872408293461,
|
1119 |
+
"grad_norm": 2.1907105445861816,
|
1120 |
+
"learning_rate": 0.00017396018370461808,
|
1121 |
+
"loss": 0.924,
|
1122 |
+
"step": 790
|
1123 |
+
},
|
1124 |
+
{
|
1125 |
+
"epoch": 0.507177033492823,
|
1126 |
+
"grad_norm": 3.8603427410125732,
|
1127 |
+
"learning_rate": 0.00017230912749291547,
|
1128 |
+
"loss": 0.9352,
|
1129 |
+
"step": 795
|
1130 |
+
},
|
1131 |
+
{
|
1132 |
+
"epoch": 0.5103668261562998,
|
1133 |
+
"grad_norm": 1.6749746799468994,
|
1134 |
+
"learning_rate": 0.00017065530255748557,
|
1135 |
+
"loss": 0.9218,
|
1136 |
+
"step": 800
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 0.5135566188197768,
|
1140 |
+
"grad_norm": 1.847925066947937,
|
1141 |
+
"learning_rate": 0.00016899891414993854,
|
1142 |
+
"loss": 0.9171,
|
1143 |
+
"step": 805
|
1144 |
+
},
|
1145 |
+
{
|
1146 |
+
"epoch": 0.5167464114832536,
|
1147 |
+
"grad_norm": 1.3573256731033325,
|
1148 |
+
"learning_rate": 0.00016734016784002994,
|
1149 |
+
"loss": 0.9178,
|
1150 |
+
"step": 810
|
1151 |
+
},
|
1152 |
+
{
|
1153 |
+
"epoch": 0.5199362041467305,
|
1154 |
+
"grad_norm": 1.2376315593719482,
|
1155 |
+
"learning_rate": 0.000165679269490148,
|
1156 |
+
"loss": 0.8975,
|
1157 |
+
"step": 815
|
1158 |
+
},
|
1159 |
+
{
|
1160 |
+
"epoch": 0.5231259968102073,
|
1161 |
+
"grad_norm": 1.7870066165924072,
|
1162 |
+
"learning_rate": 0.0001640164252297648,
|
1163 |
+
"loss": 0.9074,
|
1164 |
+
"step": 820
|
1165 |
+
},
|
1166 |
+
{
|
1167 |
+
"epoch": 0.5263157894736842,
|
1168 |
+
"grad_norm": 1.1662150621414185,
|
1169 |
+
"learning_rate": 0.00016235184142985367,
|
1170 |
+
"loss": 0.894,
|
1171 |
+
"step": 825
|
1172 |
+
},
|
1173 |
+
{
|
1174 |
+
"epoch": 0.529505582137161,
|
1175 |
+
"grad_norm": 1.8673336505889893,
|
1176 |
+
"learning_rate": 0.00016068572467727762,
|
1177 |
+
"loss": 0.9195,
|
1178 |
+
"step": 830
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 0.532695374800638,
|
1182 |
+
"grad_norm": 1.196296215057373,
|
1183 |
+
"learning_rate": 0.00015901828174915005,
|
1184 |
+
"loss": 0.9135,
|
1185 |
+
"step": 835
|
1186 |
+
},
|
1187 |
+
{
|
1188 |
+
"epoch": 0.5358851674641149,
|
1189 |
+
"grad_norm": 2.296478509902954,
|
1190 |
+
"learning_rate": 0.00015734971958717228,
|
1191 |
+
"loss": 0.9105,
|
1192 |
+
"step": 840
|
1193 |
+
},
|
1194 |
+
{
|
1195 |
+
"epoch": 0.5390749601275917,
|
1196 |
+
"grad_norm": 1.0914058685302734,
|
1197 |
+
"learning_rate": 0.00015568024527195067,
|
1198 |
+
"loss": 0.9211,
|
1199 |
+
"step": 845
|
1200 |
+
},
|
1201 |
+
{
|
1202 |
+
"epoch": 0.5422647527910686,
|
1203 |
+
"grad_norm": 1.104638934135437,
|
1204 |
+
"learning_rate": 0.0001540100659972963,
|
1205 |
+
"loss": 0.8852,
|
1206 |
+
"step": 850
|
1207 |
+
},
|
1208 |
+
{
|
1209 |
+
"epoch": 0.5454545454545454,
|
1210 |
+
"grad_norm": 1.2603851556777954,
|
1211 |
+
"learning_rate": 0.000152339389044511,
|
1212 |
+
"loss": 0.9166,
|
1213 |
+
"step": 855
|
1214 |
+
},
|
1215 |
+
{
|
1216 |
+
"epoch": 0.5486443381180224,
|
1217 |
+
"grad_norm": 1.6585578918457031,
|
1218 |
+
"learning_rate": 0.00015066842175666186,
|
1219 |
+
"loss": 0.9094,
|
1220 |
+
"step": 860
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 0.5518341307814992,
|
1224 |
+
"grad_norm": 1.1676634550094604,
|
1225 |
+
"learning_rate": 0.0001489973715128487,
|
1226 |
+
"loss": 0.8908,
|
1227 |
+
"step": 865
|
1228 |
+
},
|
1229 |
+
{
|
1230 |
+
"epoch": 0.5550239234449761,
|
1231 |
+
"grad_norm": 3.5736377239227295,
|
1232 |
+
"learning_rate": 0.00014732644570246675,
|
1233 |
+
"loss": 0.8909,
|
1234 |
+
"step": 870
|
1235 |
+
},
|
1236 |
+
{
|
1237 |
+
"epoch": 0.5582137161084529,
|
1238 |
+
"grad_norm": 1.4160521030426025,
|
1239 |
+
"learning_rate": 0.00014565585169946824,
|
1240 |
+
"loss": 0.9092,
|
1241 |
+
"step": 875
|
1242 |
+
},
|
1243 |
+
{
|
1244 |
+
"epoch": 0.5614035087719298,
|
1245 |
+
"grad_norm": 1.3239599466323853,
|
1246 |
+
"learning_rate": 0.0001439857968366256,
|
1247 |
+
"loss": 0.9054,
|
1248 |
+
"step": 880
|
1249 |
+
},
|
1250 |
+
{
|
1251 |
+
"epoch": 0.5645933014354066,
|
1252 |
+
"grad_norm": 1.075803518295288,
|
1253 |
+
"learning_rate": 0.00014231648837980022,
|
1254 |
+
"loss": 0.8836,
|
1255 |
+
"step": 885
|
1256 |
+
},
|
1257 |
+
{
|
1258 |
+
"epoch": 0.5677830940988836,
|
1259 |
+
"grad_norm": 1.0712918043136597,
|
1260 |
+
"learning_rate": 0.00014064813350221894,
|
1261 |
+
"loss": 0.9306,
|
1262 |
+
"step": 890
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 0.5709728867623605,
|
1266 |
+
"grad_norm": 1.9769309759140015,
|
1267 |
+
"learning_rate": 0.00013898093925876267,
|
1268 |
+
"loss": 0.8789,
|
1269 |
+
"step": 895
|
1270 |
+
},
|
1271 |
+
{
|
1272 |
+
"epoch": 0.5741626794258373,
|
1273 |
+
"grad_norm": 3.4712631702423096,
|
1274 |
+
"learning_rate": 0.00013731511256026913,
|
1275 |
+
"loss": 0.8883,
|
1276 |
+
"step": 900
|
1277 |
+
},
|
1278 |
+
{
|
1279 |
+
"epoch": 0.5773524720893142,
|
1280 |
+
"grad_norm": 1.1367384195327759,
|
1281 |
+
"learning_rate": 0.00013565086014785406,
|
1282 |
+
"loss": 0.908,
|
1283 |
+
"step": 905
|
1284 |
+
},
|
1285 |
+
{
|
1286 |
+
"epoch": 0.580542264752791,
|
1287 |
+
"grad_norm": 2.3845860958099365,
|
1288 |
+
"learning_rate": 0.00013398838856725257,
|
1289 |
+
"loss": 0.8968,
|
1290 |
+
"step": 910
|
1291 |
+
},
|
1292 |
+
{
|
1293 |
+
"epoch": 0.583732057416268,
|
1294 |
+
"grad_norm": 1.257691502571106,
|
1295 |
+
"learning_rate": 0.00013232790414318608,
|
1296 |
+
"loss": 0.8906,
|
1297 |
+
"step": 915
|
1298 |
+
},
|
1299 |
+
{
|
1300 |
+
"epoch": 0.5869218500797448,
|
1301 |
+
"grad_norm": 1.0229823589324951,
|
1302 |
+
"learning_rate": 0.0001306696129537553,
|
1303 |
+
"loss": 0.9062,
|
1304 |
+
"step": 920
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 0.5901116427432217,
|
1308 |
+
"grad_norm": 1.5208640098571777,
|
1309 |
+
"learning_rate": 0.00012901372080486472,
|
1310 |
+
"loss": 0.8854,
|
1311 |
+
"step": 925
|
1312 |
+
},
|
1313 |
+
{
|
1314 |
+
"epoch": 0.5933014354066986,
|
1315 |
+
"grad_norm": 1.2354867458343506,
|
1316 |
+
"learning_rate": 0.00012736043320468073,
|
1317 |
+
"loss": 0.8775,
|
1318 |
+
"step": 930
|
1319 |
+
},
|
1320 |
+
{
|
1321 |
+
"epoch": 0.5964912280701754,
|
1322 |
+
"grad_norm": 1.5079673528671265,
|
1323 |
+
"learning_rate": 0.0001257099553381262,
|
1324 |
+
"loss": 0.8893,
|
1325 |
+
"step": 935
|
1326 |
+
},
|
1327 |
+
{
|
1328 |
+
"epoch": 0.5996810207336523,
|
1329 |
+
"grad_norm": 0.9733923673629761,
|
1330 |
+
"learning_rate": 0.00012406249204141603,
|
1331 |
+
"loss": 0.89,
|
1332 |
+
"step": 940
|
1333 |
+
},
|
1334 |
+
{
|
1335 |
+
"epoch": 0.6028708133971292,
|
1336 |
+
"grad_norm": 1.3117018938064575,
|
1337 |
+
"learning_rate": 0.000122418247776635,
|
1338 |
+
"loss": 0.8945,
|
1339 |
+
"step": 945
|
1340 |
+
},
|
1341 |
+
{
|
1342 |
+
"epoch": 0.6060606060606061,
|
1343 |
+
"grad_norm": 1.251587152481079,
|
1344 |
+
"learning_rate": 0.00012077742660636299,
|
1345 |
+
"loss": 0.8909,
|
1346 |
+
"step": 950
|
1347 |
+
},
|
1348 |
+
{
|
1349 |
+
"epoch": 0.6092503987240829,
|
1350 |
+
"grad_norm": 1.1672098636627197,
|
1351 |
+
"learning_rate": 0.00011914023216834904,
|
1352 |
+
"loss": 0.8688,
|
1353 |
+
"step": 955
|
1354 |
+
},
|
1355 |
+
{
|
1356 |
+
"epoch": 0.6124401913875598,
|
1357 |
+
"grad_norm": 1.3299888372421265,
|
1358 |
+
"learning_rate": 0.0001175068676502386,
|
1359 |
+
"loss": 0.8639,
|
1360 |
+
"step": 960
|
1361 |
+
},
|
1362 |
+
{
|
1363 |
+
"epoch": 0.6156299840510366,
|
1364 |
+
"grad_norm": 1.0402354001998901,
|
1365 |
+
"learning_rate": 0.00011587753576435634,
|
1366 |
+
"loss": 0.8688,
|
1367 |
+
"step": 965
|
1368 |
+
},
|
1369 |
+
{
|
1370 |
+
"epoch": 0.6188197767145136,
|
1371 |
+
"grad_norm": 1.5155375003814697,
|
1372 |
+
"learning_rate": 0.00011425243872254835,
|
1373 |
+
"loss": 0.8738,
|
1374 |
+
"step": 970
|
1375 |
+
},
|
1376 |
+
{
|
1377 |
+
"epoch": 0.6220095693779905,
|
1378 |
+
"grad_norm": 1.1635366678237915,
|
1379 |
+
"learning_rate": 0.00011263177821108573,
|
1380 |
+
"loss": 0.8845,
|
1381 |
+
"step": 975
|
1382 |
+
},
|
1383 |
+
{
|
1384 |
+
"epoch": 0.6251993620414673,
|
1385 |
+
"grad_norm": 1.1347579956054688,
|
1386 |
+
"learning_rate": 0.00011101575536563433,
|
1387 |
+
"loss": 0.8801,
|
1388 |
+
"step": 980
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 0.6283891547049442,
|
1392 |
+
"grad_norm": 1.443695306777954,
|
1393 |
+
"learning_rate": 0.00010940457074629217,
|
1394 |
+
"loss": 0.8773,
|
1395 |
+
"step": 985
|
1396 |
+
},
|
1397 |
+
{
|
1398 |
+
"epoch": 0.631578947368421,
|
1399 |
+
"grad_norm": 1.2533496618270874,
|
1400 |
+
"learning_rate": 0.00010779842431269843,
|
1401 |
+
"loss": 0.8851,
|
1402 |
+
"step": 990
|
1403 |
+
},
|
1404 |
+
{
|
1405 |
+
"epoch": 0.6347687400318979,
|
1406 |
+
"grad_norm": 1.0504379272460938,
|
1407 |
+
"learning_rate": 0.0001061975153992172,
|
1408 |
+
"loss": 0.869,
|
1409 |
+
"step": 995
|
1410 |
+
},
|
1411 |
+
{
|
1412 |
+
"epoch": 0.6379585326953748,
|
1413 |
+
"grad_norm": 1.9185314178466797,
|
1414 |
+
"learning_rate": 0.00010460204269019829,
|
1415 |
+
"loss": 0.8623,
|
1416 |
+
"step": 1000
|
1417 |
+
},
|
1418 |
+
{
|
1419 |
+
"epoch": 0.6411483253588517,
|
1420 |
+
"grad_norm": 1.520933747291565,
|
1421 |
+
"learning_rate": 0.0001030122041953196,
|
1422 |
+
"loss": 0.8854,
|
1423 |
+
"step": 1005
|
1424 |
+
},
|
1425 |
+
{
|
1426 |
+
"epoch": 0.6443381180223285,
|
1427 |
+
"grad_norm": 1.1053557395935059,
|
1428 |
+
"learning_rate": 0.0001014281972250121,
|
1429 |
+
"loss": 0.881,
|
1430 |
+
"step": 1010
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 0.6475279106858054,
|
1434 |
+
"grad_norm": 1.258350133895874,
|
1435 |
+
"learning_rate": 9.985021836597273e-05,
|
1436 |
+
"loss": 0.8561,
|
1437 |
+
"step": 1015
|
1438 |
+
},
|
1439 |
+
{
|
1440 |
+
"epoch": 0.6507177033492823,
|
1441 |
+
"grad_norm": 1.004818081855774,
|
1442 |
+
"learning_rate": 9.827846345676614e-05,
|
1443 |
+
"loss": 0.8673,
|
1444 |
+
"step": 1020
|
1445 |
+
},
|
1446 |
+
{
|
1447 |
+
"epoch": 0.6539074960127592,
|
1448 |
+
"grad_norm": 1.3916617631912231,
|
1449 |
+
"learning_rate": 9.671312756351998e-05,
|
1450 |
+
"loss": 0.8673,
|
1451 |
+
"step": 1025
|
1452 |
+
},
|
1453 |
+
{
|
1454 |
+
"epoch": 0.6570972886762361,
|
1455 |
+
"grad_norm": 1.0740200281143188,
|
1456 |
+
"learning_rate": 9.515440495571569e-05,
|
1457 |
+
"loss": 0.8678,
|
1458 |
+
"step": 1030
|
1459 |
+
},
|
1460 |
+
{
|
1461 |
+
"epoch": 0.6602870813397129,
|
1462 |
+
"grad_norm": 1.025354266166687,
|
1463 |
+
"learning_rate": 9.360248908207813e-05,
|
1464 |
+
"loss": 0.8611,
|
1465 |
+
"step": 1035
|
1466 |
+
},
|
1467 |
+
{
|
1468 |
+
"epoch": 0.6634768740031898,
|
1469 |
+
"grad_norm": 1.2619115114212036,
|
1470 |
+
"learning_rate": 9.205757254656755e-05,
|
1471 |
+
"loss": 0.8485,
|
1472 |
+
"step": 1040
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 0.6666666666666666,
|
1476 |
+
"grad_norm": 1.9425147771835327,
|
1477 |
+
"learning_rate": 9.05198470844756e-05,
|
1478 |
+
"loss": 0.8534,
|
1479 |
+
"step": 1045
|
1480 |
+
},
|
1481 |
+
{
|
1482 |
+
"epoch": 0.6698564593301436,
|
1483 |
+
"grad_norm": 1.214699149131775,
|
1484 |
+
"learning_rate": 8.898950353862998e-05,
|
1485 |
+
"loss": 0.8739,
|
1486 |
+
"step": 1050
|
1487 |
+
},
|
1488 |
+
{
|
1489 |
+
"epoch": 0.6730462519936204,
|
1490 |
+
"grad_norm": 1.2433393001556396,
|
1491 |
+
"learning_rate": 8.746673183570923e-05,
|
1492 |
+
"loss": 0.8703,
|
1493 |
+
"step": 1055
|
1494 |
+
},
|
1495 |
+
{
|
1496 |
+
"epoch": 0.6762360446570973,
|
1497 |
+
"grad_norm": 1.4109057188034058,
|
1498 |
+
"learning_rate": 8.595172096267157e-05,
|
1499 |
+
"loss": 0.9019,
|
1500 |
+
"step": 1060
|
1501 |
+
},
|
1502 |
+
{
|
1503 |
+
"epoch": 0.6794258373205742,
|
1504 |
+
"grad_norm": 1.2065266370773315,
|
1505 |
+
"learning_rate": 8.444465894330024e-05,
|
1506 |
+
"loss": 0.8465,
|
1507 |
+
"step": 1065
|
1508 |
+
},
|
1509 |
+
{
|
1510 |
+
"epoch": 0.682615629984051,
|
1511 |
+
"grad_norm": 0.994489848613739,
|
1512 |
+
"learning_rate": 8.294573281486828e-05,
|
1513 |
+
"loss": 0.8654,
|
1514 |
+
"step": 1070
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 0.6858054226475279,
|
1518 |
+
"grad_norm": 1.036932110786438,
|
1519 |
+
"learning_rate": 8.145512860492596e-05,
|
1520 |
+
"loss": 0.8509,
|
1521 |
+
"step": 1075
|
1522 |
+
},
|
1523 |
+
{
|
1524 |
+
"epoch": 0.6889952153110048,
|
1525 |
+
"grad_norm": 1.1829875707626343,
|
1526 |
+
"learning_rate": 7.997303130821362e-05,
|
1527 |
+
"loss": 0.8726,
|
1528 |
+
"step": 1080
|
1529 |
+
},
|
1530 |
+
{
|
1531 |
+
"epoch": 0.6921850079744817,
|
1532 |
+
"grad_norm": 1.0227490663528442,
|
1533 |
+
"learning_rate": 7.849962486370206e-05,
|
1534 |
+
"loss": 0.8623,
|
1535 |
+
"step": 1085
|
1536 |
+
},
|
1537 |
+
{
|
1538 |
+
"epoch": 0.6953748006379585,
|
1539 |
+
"grad_norm": 1.0316836833953857,
|
1540 |
+
"learning_rate": 7.703509213176451e-05,
|
1541 |
+
"loss": 0.8762,
|
1542 |
+
"step": 1090
|
1543 |
+
},
|
1544 |
+
{
|
1545 |
+
"epoch": 0.6985645933014354,
|
1546 |
+
"grad_norm": 1.1211485862731934,
|
1547 |
+
"learning_rate": 7.557961487148272e-05,
|
1548 |
+
"loss": 0.8824,
|
1549 |
+
"step": 1095
|
1550 |
+
},
|
1551 |
+
{
|
1552 |
+
"epoch": 0.7017543859649122,
|
1553 |
+
"grad_norm": 1.0168806314468384,
|
1554 |
+
"learning_rate": 7.413337371808884e-05,
|
1555 |
+
"loss": 0.8582,
|
1556 |
+
"step": 1100
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 0.7049441786283892,
|
1560 |
+
"grad_norm": 1.244272232055664,
|
1561 |
+
"learning_rate": 7.269654816054756e-05,
|
1562 |
+
"loss": 0.8563,
|
1563 |
+
"step": 1105
|
1564 |
+
},
|
1565 |
+
{
|
1566 |
+
"epoch": 0.7081339712918661,
|
1567 |
+
"grad_norm": 1.1271004676818848,
|
1568 |
+
"learning_rate": 7.126931651928012e-05,
|
1569 |
+
"loss": 0.8538,
|
1570 |
+
"step": 1110
|
1571 |
+
},
|
1572 |
+
{
|
1573 |
+
"epoch": 0.7113237639553429,
|
1574 |
+
"grad_norm": 1.1314244270324707,
|
1575 |
+
"learning_rate": 6.985185592403367e-05,
|
1576 |
+
"loss": 0.8716,
|
1577 |
+
"step": 1115
|
1578 |
+
},
|
1579 |
+
{
|
1580 |
+
"epoch": 0.7145135566188198,
|
1581 |
+
"grad_norm": 1.212408423423767,
|
1582 |
+
"learning_rate": 6.844434229189787e-05,
|
1583 |
+
"loss": 0.8624,
|
1584 |
+
"step": 1120
|
1585 |
+
},
|
1586 |
+
{
|
1587 |
+
"epoch": 0.7177033492822966,
|
1588 |
+
"grad_norm": 1.4100135564804077,
|
1589 |
+
"learning_rate": 6.704695030547252e-05,
|
1590 |
+
"loss": 0.8428,
|
1591 |
+
"step": 1125
|
1592 |
+
},
|
1593 |
+
{
|
1594 |
+
"epoch": 0.7208931419457735,
|
1595 |
+
"grad_norm": 1.6255112886428833,
|
1596 |
+
"learning_rate": 6.56598533911881e-05,
|
1597 |
+
"loss": 0.8553,
|
1598 |
+
"step": 1130
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 0.7240829346092504,
|
1602 |
+
"grad_norm": 1.0644069910049438,
|
1603 |
+
"learning_rate": 6.428322369778254e-05,
|
1604 |
+
"loss": 0.8703,
|
1605 |
+
"step": 1135
|
1606 |
+
},
|
1607 |
+
{
|
1608 |
+
"epoch": 0.7272727272727273,
|
1609 |
+
"grad_norm": 0.9541735053062439,
|
1610 |
+
"learning_rate": 6.291723207493577e-05,
|
1611 |
+
"loss": 0.8334,
|
1612 |
+
"step": 1140
|
1613 |
+
},
|
1614 |
+
{
|
1615 |
+
"epoch": 0.7304625199362041,
|
1616 |
+
"grad_norm": 1.1123586893081665,
|
1617 |
+
"learning_rate": 6.15620480520666e-05,
|
1618 |
+
"loss": 0.8521,
|
1619 |
+
"step": 1145
|
1620 |
+
},
|
1621 |
+
{
|
1622 |
+
"epoch": 0.733652312599681,
|
1623 |
+
"grad_norm": 1.3636609315872192,
|
1624 |
+
"learning_rate": 6.0217839817292765e-05,
|
1625 |
+
"loss": 0.84,
|
1626 |
+
"step": 1150
|
1627 |
+
},
|
1628 |
+
{
|
1629 |
+
"epoch": 0.7368421052631579,
|
1630 |
+
"grad_norm": 1.311625599861145,
|
1631 |
+
"learning_rate": 5.888477419655733e-05,
|
1632 |
+
"loss": 0.8592,
|
1633 |
+
"step": 1155
|
1634 |
+
},
|
1635 |
+
{
|
1636 |
+
"epoch": 0.7400318979266348,
|
1637 |
+
"grad_norm": 1.0517165660858154,
|
1638 |
+
"learning_rate": 5.7563016632924555e-05,
|
1639 |
+
"loss": 0.8403,
|
1640 |
+
"step": 1160
|
1641 |
+
},
|
1642 |
+
{
|
1643 |
+
"epoch": 0.7432216905901117,
|
1644 |
+
"grad_norm": 1.1106277704238892,
|
1645 |
+
"learning_rate": 5.625273116604728e-05,
|
1646 |
+
"loss": 0.8828,
|
1647 |
+
"step": 1165
|
1648 |
+
},
|
1649 |
+
{
|
1650 |
+
"epoch": 0.7464114832535885,
|
1651 |
+
"grad_norm": 1.4360631704330444,
|
1652 |
+
"learning_rate": 5.495408041180829e-05,
|
1653 |
+
"loss": 0.8732,
|
1654 |
+
"step": 1170
|
1655 |
+
},
|
1656 |
+
{
|
1657 |
+
"epoch": 0.7496012759170654,
|
1658 |
+
"grad_norm": 1.226717233657837,
|
1659 |
+
"learning_rate": 5.3667225542138507e-05,
|
1660 |
+
"loss": 0.8268,
|
1661 |
+
"step": 1175
|
1662 |
+
},
|
1663 |
+
{
|
1664 |
+
"epoch": 0.7527910685805422,
|
1665 |
+
"grad_norm": 1.013719916343689,
|
1666 |
+
"learning_rate": 5.239232626501464e-05,
|
1667 |
+
"loss": 0.8775,
|
1668 |
+
"step": 1180
|
1669 |
+
},
|
1670 |
+
{
|
1671 |
+
"epoch": 0.7559808612440191,
|
1672 |
+
"grad_norm": 1.9293391704559326,
|
1673 |
+
"learning_rate": 5.1129540804637906e-05,
|
1674 |
+
"loss": 0.8578,
|
1675 |
+
"step": 1185
|
1676 |
+
},
|
1677 |
+
{
|
1678 |
+
"epoch": 0.759170653907496,
|
1679 |
+
"grad_norm": 1.105729103088379,
|
1680 |
+
"learning_rate": 4.9879025881797466e-05,
|
1681 |
+
"loss": 0.8443,
|
1682 |
+
"step": 1190
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 0.7623604465709729,
|
1686 |
+
"grad_norm": 1.0743812322616577,
|
1687 |
+
"learning_rate": 4.8640936694420245e-05,
|
1688 |
+
"loss": 0.8406,
|
1689 |
+
"step": 1195
|
1690 |
+
},
|
1691 |
+
{
|
1692 |
+
"epoch": 0.7655502392344498,
|
1693 |
+
"grad_norm": 1.0439478158950806,
|
1694 |
+
"learning_rate": 4.7415426898309704e-05,
|
1695 |
+
"loss": 0.8281,
|
1696 |
+
"step": 1200
|
1697 |
+
},
|
1698 |
+
{
|
1699 |
+
"epoch": 0.7687400318979266,
|
1700 |
+
"grad_norm": 1.1847258806228638,
|
1701 |
+
"learning_rate": 4.62026485880761e-05,
|
1702 |
+
"loss": 0.845,
|
1703 |
+
"step": 1205
|
1704 |
+
},
|
1705 |
+
{
|
1706 |
+
"epoch": 0.7719298245614035,
|
1707 |
+
"grad_norm": 1.4025248289108276,
|
1708 |
+
"learning_rate": 4.50027522782603e-05,
|
1709 |
+
"loss": 0.8456,
|
1710 |
+
"step": 1210
|
1711 |
+
},
|
1712 |
+
{
|
1713 |
+
"epoch": 0.7751196172248804,
|
1714 |
+
"grad_norm": 1.398414134979248,
|
1715 |
+
"learning_rate": 4.3815886884654136e-05,
|
1716 |
+
"loss": 0.8367,
|
1717 |
+
"step": 1215
|
1718 |
+
},
|
1719 |
+
{
|
1720 |
+
"epoch": 0.7783094098883573,
|
1721 |
+
"grad_norm": 1.4555211067199707,
|
1722 |
+
"learning_rate": 4.264219970581854e-05,
|
1723 |
+
"loss": 0.8594,
|
1724 |
+
"step": 1220
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 0.7814992025518341,
|
1728 |
+
"grad_norm": 1.2245261669158936,
|
1729 |
+
"learning_rate": 4.148183640480293e-05,
|
1730 |
+
"loss": 0.8622,
|
1731 |
+
"step": 1225
|
1732 |
+
},
|
1733 |
+
{
|
1734 |
+
"epoch": 0.784688995215311,
|
1735 |
+
"grad_norm": 1.0004338026046753,
|
1736 |
+
"learning_rate": 4.0334940991067276e-05,
|
1737 |
+
"loss": 0.8619,
|
1738 |
+
"step": 1230
|
1739 |
+
},
|
1740 |
+
{
|
1741 |
+
"epoch": 0.7878787878787878,
|
1742 |
+
"grad_norm": 0.9635859131813049,
|
1743 |
+
"learning_rate": 3.920165580260973e-05,
|
1744 |
+
"loss": 0.8596,
|
1745 |
+
"step": 1235
|
1746 |
+
},
|
1747 |
+
{
|
1748 |
+
"epoch": 0.7910685805422647,
|
1749 |
+
"grad_norm": 1.0776612758636475,
|
1750 |
+
"learning_rate": 3.808212148830095e-05,
|
1751 |
+
"loss": 0.839,
|
1752 |
+
"step": 1240
|
1753 |
+
},
|
1754 |
+
{
|
1755 |
+
"epoch": 0.7942583732057417,
|
1756 |
+
"grad_norm": 1.1850311756134033,
|
1757 |
+
"learning_rate": 3.697647699042918e-05,
|
1758 |
+
"loss": 0.8347,
|
1759 |
+
"step": 1245
|
1760 |
+
},
|
1761 |
+
{
|
1762 |
+
"epoch": 0.7974481658692185,
|
1763 |
+
"grad_norm": 1.168291449546814,
|
1764 |
+
"learning_rate": 3.5884859527455995e-05,
|
1765 |
+
"loss": 0.8513,
|
1766 |
+
"step": 1250
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 0.8006379585326954,
|
1770 |
+
"grad_norm": 1.0602375268936157,
|
1771 |
+
"learning_rate": 3.48074045769868e-05,
|
1772 |
+
"loss": 0.8521,
|
1773 |
+
"step": 1255
|
1774 |
+
},
|
1775 |
+
{
|
1776 |
+
"epoch": 0.8038277511961722,
|
1777 |
+
"grad_norm": 1.074064016342163,
|
1778 |
+
"learning_rate": 3.3744245858956967e-05,
|
1779 |
+
"loss": 0.877,
|
1780 |
+
"step": 1260
|
1781 |
+
},
|
1782 |
+
{
|
1783 |
+
"epoch": 0.8070175438596491,
|
1784 |
+
"grad_norm": 1.0063201189041138,
|
1785 |
+
"learning_rate": 3.2695515319036186e-05,
|
1786 |
+
"loss": 0.8549,
|
1787 |
+
"step": 1265
|
1788 |
+
},
|
1789 |
+
{
|
1790 |
+
"epoch": 0.810207336523126,
|
1791 |
+
"grad_norm": 0.9187350273132324,
|
1792 |
+
"learning_rate": 3.1661343112253304e-05,
|
1793 |
+
"loss": 0.8439,
|
1794 |
+
"step": 1270
|
1795 |
+
},
|
1796 |
+
{
|
1797 |
+
"epoch": 0.8133971291866029,
|
1798 |
+
"grad_norm": 1.0719003677368164,
|
1799 |
+
"learning_rate": 3.064185758684265e-05,
|
1800 |
+
"loss": 0.829,
|
1801 |
+
"step": 1275
|
1802 |
+
},
|
1803 |
+
{
|
1804 |
+
"epoch": 0.8165869218500797,
|
1805 |
+
"grad_norm": 0.9533995985984802,
|
1806 |
+
"learning_rate": 2.96371852683157e-05,
|
1807 |
+
"loss": 0.8492,
|
1808 |
+
"step": 1280
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 0.8197767145135566,
|
1812 |
+
"grad_norm": 1.3275169134140015,
|
1813 |
+
"learning_rate": 2.8647450843757897e-05,
|
1814 |
+
"loss": 0.8587,
|
1815 |
+
"step": 1285
|
1816 |
+
},
|
1817 |
+
{
|
1818 |
+
"epoch": 0.8229665071770335,
|
1819 |
+
"grad_norm": 1.1175739765167236,
|
1820 |
+
"learning_rate": 2.7672777146354246e-05,
|
1821 |
+
"loss": 0.8505,
|
1822 |
+
"step": 1290
|
1823 |
+
},
|
1824 |
+
{
|
1825 |
+
"epoch": 0.8261562998405104,
|
1826 |
+
"grad_norm": 1.0937777757644653,
|
1827 |
+
"learning_rate": 2.6713285140144802e-05,
|
1828 |
+
"loss": 0.8624,
|
1829 |
+
"step": 1295
|
1830 |
+
},
|
1831 |
+
{
|
1832 |
+
"epoch": 0.8293460925039873,
|
1833 |
+
"grad_norm": 1.5751278400421143,
|
1834 |
+
"learning_rate": 2.5769093905012333e-05,
|
1835 |
+
"loss": 0.8406,
|
1836 |
+
"step": 1300
|
1837 |
+
},
|
1838 |
+
{
|
1839 |
+
"epoch": 0.8325358851674641,
|
1840 |
+
"grad_norm": 0.9947794079780579,
|
1841 |
+
"learning_rate": 2.4840320621903253e-05,
|
1842 |
+
"loss": 0.8374,
|
1843 |
+
"step": 1305
|
1844 |
+
},
|
1845 |
+
{
|
1846 |
+
"epoch": 0.835725677830941,
|
1847 |
+
"grad_norm": 1.0281869173049927,
|
1848 |
+
"learning_rate": 2.392708055828495e-05,
|
1849 |
+
"loss": 0.8417,
|
1850 |
+
"step": 1310
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 0.8389154704944178,
|
1854 |
+
"grad_norm": 1.0307683944702148,
|
1855 |
+
"learning_rate": 2.3029487053840295e-05,
|
1856 |
+
"loss": 0.8379,
|
1857 |
+
"step": 1315
|
1858 |
+
},
|
1859 |
+
{
|
1860 |
+
"epoch": 0.8421052631578947,
|
1861 |
+
"grad_norm": 0.9527477025985718,
|
1862 |
+
"learning_rate": 2.214765150640108e-05,
|
1863 |
+
"loss": 0.8518,
|
1864 |
+
"step": 1320
|
1865 |
+
},
|
1866 |
+
{
|
1867 |
+
"epoch": 0.8452950558213717,
|
1868 |
+
"grad_norm": 1.0322529077529907,
|
1869 |
+
"learning_rate": 2.1281683358122996e-05,
|
1870 |
+
"loss": 0.8398,
|
1871 |
+
"step": 1325
|
1872 |
+
},
|
1873 |
+
{
|
1874 |
+
"epoch": 0.8484848484848485,
|
1875 |
+
"grad_norm": 0.9636293053627014,
|
1876 |
+
"learning_rate": 2.043169008190289e-05,
|
1877 |
+
"loss": 0.859,
|
1878 |
+
"step": 1330
|
1879 |
+
},
|
1880 |
+
{
|
1881 |
+
"epoch": 0.8516746411483254,
|
1882 |
+
"grad_norm": 0.946626603603363,
|
1883 |
+
"learning_rate": 1.9597777168040872e-05,
|
1884 |
+
"loss": 0.8586,
|
1885 |
+
"step": 1335
|
1886 |
+
},
|
1887 |
+
{
|
1888 |
+
"epoch": 0.8548644338118022,
|
1889 |
+
"grad_norm": 0.9317569136619568,
|
1890 |
+
"learning_rate": 1.8780048111147776e-05,
|
1891 |
+
"loss": 0.8587,
|
1892 |
+
"step": 1340
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 0.8580542264752791,
|
1896 |
+
"grad_norm": 1.0601786375045776,
|
1897 |
+
"learning_rate": 1.797860439730126e-05,
|
1898 |
+
"loss": 0.8097,
|
1899 |
+
"step": 1345
|
1900 |
+
},
|
1901 |
+
{
|
1902 |
+
"epoch": 0.861244019138756,
|
1903 |
+
"grad_norm": 0.9538899064064026,
|
1904 |
+
"learning_rate": 1.7193545491450183e-05,
|
1905 |
+
"loss": 0.8533,
|
1906 |
+
"step": 1350
|
1907 |
+
},
|
1908 |
+
{
|
1909 |
+
"epoch": 0.8644338118022329,
|
1910 |
+
"grad_norm": 0.9394316077232361,
|
1911 |
+
"learning_rate": 1.6424968825070567e-05,
|
1912 |
+
"loss": 0.8271,
|
1913 |
+
"step": 1355
|
1914 |
+
},
|
1915 |
+
{
|
1916 |
+
"epoch": 0.8676236044657097,
|
1917 |
+
"grad_norm": 1.0136967897415161,
|
1918 |
+
"learning_rate": 1.567296978407353e-05,
|
1919 |
+
"loss": 0.8378,
|
1920 |
+
"step": 1360
|
1921 |
+
},
|
1922 |
+
{
|
1923 |
+
"epoch": 0.8708133971291866,
|
1924 |
+
"grad_norm": 0.9502097964286804,
|
1925 |
+
"learning_rate": 1.4937641696967245e-05,
|
1926 |
+
"loss": 0.8568,
|
1927 |
+
"step": 1365
|
1928 |
+
},
|
1929 |
+
{
|
1930 |
+
"epoch": 0.8740031897926634,
|
1931 |
+
"grad_norm": 0.9635610580444336,
|
1932 |
+
"learning_rate": 1.4219075823274251e-05,
|
1933 |
+
"loss": 0.8426,
|
1934 |
+
"step": 1370
|
1935 |
+
},
|
1936 |
+
{
|
1937 |
+
"epoch": 0.8771929824561403,
|
1938 |
+
"grad_norm": 0.9333298206329346,
|
1939 |
+
"learning_rate": 1.3517361342205295e-05,
|
1940 |
+
"loss": 0.8346,
|
1941 |
+
"step": 1375
|
1942 |
+
},
|
1943 |
+
{
|
1944 |
+
"epoch": 0.8803827751196173,
|
1945 |
+
"grad_norm": 0.874499499797821,
|
1946 |
+
"learning_rate": 1.2832585341591844e-05,
|
1947 |
+
"loss": 0.8138,
|
1948 |
+
"step": 1380
|
1949 |
+
},
|
1950 |
+
{
|
1951 |
+
"epoch": 0.8835725677830941,
|
1952 |
+
"grad_norm": 1.1882647275924683,
|
1953 |
+
"learning_rate": 1.2164832807077585e-05,
|
1954 |
+
"loss": 0.8446,
|
1955 |
+
"step": 1385
|
1956 |
+
},
|
1957 |
+
{
|
1958 |
+
"epoch": 0.886762360446571,
|
1959 |
+
"grad_norm": 1.0986793041229248,
|
1960 |
+
"learning_rate": 1.151418661157122e-05,
|
1961 |
+
"loss": 0.8343,
|
1962 |
+
"step": 1390
|
1963 |
+
},
|
1964 |
+
{
|
1965 |
+
"epoch": 0.8899521531100478,
|
1966 |
+
"grad_norm": 0.930297315120697,
|
1967 |
+
"learning_rate": 1.0880727504961339e-05,
|
1968 |
+
"loss": 0.8307,
|
1969 |
+
"step": 1395
|
1970 |
+
},
|
1971 |
+
{
|
1972 |
+
"epoch": 0.8931419457735247,
|
1973 |
+
"grad_norm": 0.9573615193367004,
|
1974 |
+
"learning_rate": 1.0264534104094812e-05,
|
1975 |
+
"loss": 0.8423,
|
1976 |
+
"step": 1400
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 0.8963317384370016,
|
1980 |
+
"grad_norm": 0.976313591003418,
|
1981 |
+
"learning_rate": 9.665682883019732e-06,
|
1982 |
+
"loss": 0.833,
|
1983 |
+
"step": 1405
|
1984 |
+
},
|
1985 |
+
{
|
1986 |
+
"epoch": 0.8995215311004785,
|
1987 |
+
"grad_norm": 0.9044297337532043,
|
1988 |
+
"learning_rate": 9.084248163494511e-06,
|
1989 |
+
"loss": 0.838,
|
1990 |
+
"step": 1410
|
1991 |
+
},
|
1992 |
+
{
|
1993 |
+
"epoch": 0.9027113237639554,
|
1994 |
+
"grad_norm": 0.9953575134277344,
|
1995 |
+
"learning_rate": 8.520302105764148e-06,
|
1996 |
+
"loss": 0.8357,
|
1997 |
+
"step": 1415
|
1998 |
+
},
|
1999 |
+
{
|
2000 |
+
"epoch": 0.9059011164274322,
|
2001 |
+
"grad_norm": 1.0662360191345215,
|
2002 |
+
"learning_rate": 7.973914699604367e-06,
|
2003 |
+
"loss": 0.8355,
|
2004 |
+
"step": 1420
|
2005 |
+
},
|
2006 |
+
{
|
2007 |
+
"epoch": 0.9090909090909091,
|
2008 |
+
"grad_norm": 0.9103978276252747,
|
2009 |
+
"learning_rate": 7.445153755635569e-06,
|
2010 |
+
"loss": 0.8451,
|
2011 |
+
"step": 1425
|
2012 |
+
},
|
2013 |
+
{
|
2014 |
+
"epoch": 0.9122807017543859,
|
2015 |
+
"grad_norm": 0.9437443614006042,
|
2016 |
+
"learning_rate": 6.934084896906983e-06,
|
2017 |
+
"loss": 0.8431,
|
2018 |
+
"step": 1430
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 0.9154704944178629,
|
2022 |
+
"grad_norm": 1.0551937818527222,
|
2023 |
+
"learning_rate": 6.440771550752377e-06,
|
2024 |
+
"loss": 0.8508,
|
2025 |
+
"step": 1435
|
2026 |
+
},
|
2027 |
+
{
|
2028 |
+
"epoch": 0.9186602870813397,
|
2029 |
+
"grad_norm": 0.9969338178634644,
|
2030 |
+
"learning_rate": 5.965274940918274e-06,
|
2031 |
+
"loss": 0.84,
|
2032 |
+
"step": 1440
|
2033 |
+
},
|
2034 |
+
{
|
2035 |
+
"epoch": 0.9218500797448166,
|
2036 |
+
"grad_norm": 0.9045847654342651,
|
2037 |
+
"learning_rate": 5.507654079965612e-06,
|
2038 |
+
"loss": 0.8169,
|
2039 |
+
"step": 1445
|
2040 |
+
},
|
2041 |
+
{
|
2042 |
+
"epoch": 0.9250398724082934,
|
2043 |
+
"grad_norm": 1.1404656171798706,
|
2044 |
+
"learning_rate": 5.067965761945869e-06,
|
2045 |
+
"loss": 0.8529,
|
2046 |
+
"step": 1450
|
2047 |
+
},
|
2048 |
+
{
|
2049 |
+
"epoch": 0.9282296650717703,
|
2050 |
+
"grad_norm": 0.9272449016571045,
|
2051 |
+
"learning_rate": 4.646264555352586e-06,
|
2052 |
+
"loss": 0.8292,
|
2053 |
+
"step": 1455
|
2054 |
+
},
|
2055 |
+
{
|
2056 |
+
"epoch": 0.9314194577352473,
|
2057 |
+
"grad_norm": 1.0107725858688354,
|
2058 |
+
"learning_rate": 4.242602796348915e-06,
|
2059 |
+
"loss": 0.8219,
|
2060 |
+
"step": 1460
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 0.9346092503987241,
|
2064 |
+
"grad_norm": 1.1535266637802124,
|
2065 |
+
"learning_rate": 3.857030582272369e-06,
|
2066 |
+
"loss": 0.8254,
|
2067 |
+
"step": 1465
|
2068 |
+
},
|
2069 |
+
{
|
2070 |
+
"epoch": 0.937799043062201,
|
2071 |
+
"grad_norm": 1.0003602504730225,
|
2072 |
+
"learning_rate": 3.489595765417441e-06,
|
2073 |
+
"loss": 0.7996,
|
2074 |
+
"step": 1470
|
2075 |
+
},
|
2076 |
+
{
|
2077 |
+
"epoch": 0.9409888357256778,
|
2078 |
+
"grad_norm": 0.9239038228988647,
|
2079 |
+
"learning_rate": 3.140343947096624e-06,
|
2080 |
+
"loss": 0.8079,
|
2081 |
+
"step": 1475
|
2082 |
+
},
|
2083 |
+
{
|
2084 |
+
"epoch": 0.9441786283891547,
|
2085 |
+
"grad_norm": 0.9535396695137024,
|
2086 |
+
"learning_rate": 2.80931847198117e-06,
|
2087 |
+
"loss": 0.861,
|
2088 |
+
"step": 1480
|
2089 |
+
},
|
2090 |
+
{
|
2091 |
+
"epoch": 0.9473684210526315,
|
2092 |
+
"grad_norm": 1.0079741477966309,
|
2093 |
+
"learning_rate": 2.4965604227215774e-06,
|
2094 |
+
"loss": 0.844,
|
2095 |
+
"step": 1485
|
2096 |
+
},
|
2097 |
+
{
|
2098 |
+
"epoch": 0.9505582137161085,
|
2099 |
+
"grad_norm": 0.9593281149864197,
|
2100 |
+
"learning_rate": 2.202108614848885e-06,
|
2101 |
+
"loss": 0.8295,
|
2102 |
+
"step": 1490
|
2103 |
+
},
|
2104 |
+
{
|
2105 |
+
"epoch": 0.9537480063795853,
|
2106 |
+
"grad_norm": 0.8958705067634583,
|
2107 |
+
"learning_rate": 1.925999591957561e-06,
|
2108 |
+
"loss": 0.8235,
|
2109 |
+
"step": 1495
|
2110 |
+
},
|
2111 |
+
{
|
2112 |
+
"epoch": 0.9569377990430622,
|
2113 |
+
"grad_norm": 0.9902259707450867,
|
2114 |
+
"learning_rate": 1.6682676211700107e-06,
|
2115 |
+
"loss": 0.8371,
|
2116 |
+
"step": 1500
|
2117 |
+
},
|
2118 |
+
{
|
2119 |
+
"epoch": 0.960127591706539,
|
2120 |
+
"grad_norm": 0.906568706035614,
|
2121 |
+
"learning_rate": 1.4289446888838652e-06,
|
2122 |
+
"loss": 0.8421,
|
2123 |
+
"step": 1505
|
2124 |
+
},
|
2125 |
+
{
|
2126 |
+
"epoch": 0.9633173843700159,
|
2127 |
+
"grad_norm": 1.0222581624984741,
|
2128 |
+
"learning_rate": 1.2080604968022378e-06,
|
2129 |
+
"loss": 0.8104,
|
2130 |
+
"step": 1510
|
2131 |
+
},
|
2132 |
+
{
|
2133 |
+
"epoch": 0.9665071770334929,
|
2134 |
+
"grad_norm": 0.9215661883354187,
|
2135 |
+
"learning_rate": 1.0056424582474575e-06,
|
2136 |
+
"loss": 0.8266,
|
2137 |
+
"step": 1515
|
2138 |
+
},
|
2139 |
+
{
|
2140 |
+
"epoch": 0.9696969696969697,
|
2141 |
+
"grad_norm": 0.9094756245613098,
|
2142 |
+
"learning_rate": 8.217156947590064e-07,
|
2143 |
+
"loss": 0.8195,
|
2144 |
+
"step": 1520
|
2145 |
+
},
|
2146 |
+
{
|
2147 |
+
"epoch": 0.9728867623604466,
|
2148 |
+
"grad_norm": 0.9477907419204712,
|
2149 |
+
"learning_rate": 6.563030329755969e-07,
|
2150 |
+
"loss": 0.807,
|
2151 |
+
"step": 1525
|
2152 |
+
},
|
2153 |
+
{
|
2154 |
+
"epoch": 0.9760765550239234,
|
2155 |
+
"grad_norm": 1.0699169635772705,
|
2156 |
+
"learning_rate": 5.094250018023715e-07,
|
2157 |
+
"loss": 0.8512,
|
2158 |
+
"step": 1530
|
2159 |
+
},
|
2160 |
+
{
|
2161 |
+
"epoch": 0.9792663476874003,
|
2162 |
+
"grad_norm": 0.9320189952850342,
|
2163 |
+
"learning_rate": 3.8109982986300747e-07,
|
2164 |
+
"loss": 0.8457,
|
2165 |
+
"step": 1535
|
2166 |
+
},
|
2167 |
+
{
|
2168 |
+
"epoch": 0.9824561403508771,
|
2169 |
+
"grad_norm": 0.9529619812965393,
|
2170 |
+
"learning_rate": 2.7134344323747616e-07,
|
2171 |
+
"loss": 0.8232,
|
2172 |
+
"step": 1540
|
2173 |
+
},
|
2174 |
+
{
|
2175 |
+
"epoch": 0.9856459330143541,
|
2176 |
+
"grad_norm": 0.9721714854240417,
|
2177 |
+
"learning_rate": 1.801694634854578e-07,
|
2178 |
+
"loss": 0.8438,
|
2179 |
+
"step": 1545
|
2180 |
+
},
|
2181 |
+
{
|
2182 |
+
"epoch": 0.988835725677831,
|
2183 |
+
"grad_norm": 1.0308058261871338,
|
2184 |
+
"learning_rate": 1.075892059558603e-07,
|
2185 |
+
"loss": 0.8509,
|
2186 |
+
"step": 1550
|
2187 |
+
},
|
2188 |
+
{
|
2189 |
+
"epoch": 0.9920255183413078,
|
2190 |
+
"grad_norm": 0.9877679347991943,
|
2191 |
+
"learning_rate": 5.3611678382442516e-08,
|
2192 |
+
"loss": 0.8327,
|
2193 |
+
"step": 1555
|
2194 |
+
},
|
2195 |
+
{
|
2196 |
+
"epoch": 0.9952153110047847,
|
2197 |
+
"grad_norm": 0.987597644329071,
|
2198 |
+
"learning_rate": 1.824357976594193e-08,
|
2199 |
+
"loss": 0.8036,
|
2200 |
+
"step": 1560
|
2201 |
+
},
|
2202 |
+
{
|
2203 |
+
"epoch": 0.9984051036682615,
|
2204 |
+
"grad_norm": 1.006822943687439,
|
2205 |
+
"learning_rate": 1.4892995426396548e-09,
|
2206 |
+
"loss": 0.8301,
|
2207 |
+
"step": 1565
|
2208 |
+
},
|
2209 |
+
{
|
2210 |
+
"epoch": 0.9996810207336523,
|
2211 |
+
"eval_loss": 2.3456168174743652,
|
2212 |
+
"eval_runtime": 2.9374,
|
2213 |
+
"eval_samples_per_second": 3.404,
|
2214 |
+
"eval_steps_per_second": 0.681,
|
2215 |
+
"step": 1567
|
2216 |
+
},
|
2217 |
+
{
|
2218 |
+
"epoch": 0.9996810207336523,
|
2219 |
+
"step": 1567,
|
2220 |
+
"total_flos": 6.008617478547046e+17,
|
2221 |
+
"train_loss": 1.5171029353948178,
|
2222 |
+
"train_runtime": 9750.6774,
|
2223 |
+
"train_samples_per_second": 1.286,
|
2224 |
+
"train_steps_per_second": 0.161
|
2225 |
+
}
|
2226 |
+
],
|
2227 |
+
"logging_steps": 5,
|
2228 |
+
"max_steps": 1567,
|
2229 |
+
"num_input_tokens_seen": 0,
|
2230 |
+
"num_train_epochs": 1,
|
2231 |
+
"save_steps": 100,
|
2232 |
+
"stateful_callbacks": {
|
2233 |
+
"TrainerControl": {
|
2234 |
+
"args": {
|
2235 |
+
"should_epoch_stop": false,
|
2236 |
+
"should_evaluate": false,
|
2237 |
+
"should_log": false,
|
2238 |
+
"should_save": true,
|
2239 |
+
"should_training_stop": true
|
2240 |
+
},
|
2241 |
+
"attributes": {}
|
2242 |
+
}
|
2243 |
+
},
|
2244 |
+
"total_flos": 6.008617478547046e+17,
|
2245 |
+
"train_batch_size": 1,
|
2246 |
+
"trial_name": null,
|
2247 |
+
"trial_params": null
|
2248 |
+
}
|