Commit
•
66cd223
1
Parent(s):
4eae9d0
Model save
Browse files- README.md +65 -0
- all_results.json +9 -0
- generation_config.json +11 -0
- runs/Aug02_22-17-19_poseidon/events.out.tfevents.1722637052.poseidon.1348942.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1947 -0
README.md
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gemma
|
| 3 |
+
base_model: google/gemma-2-2b-it
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: gemma-2-2b-it-sum-sft-full_1epochs
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# gemma-2-2b-it-sum-sft-full_1epochs
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [google/gemma-2-2b-it](https://huggingface.co/google/gemma-2-2b-it) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 2.1641
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 2e-05
|
| 42 |
+
- train_batch_size: 1
|
| 43 |
+
- eval_batch_size: 2
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- gradient_accumulation_steps: 16
|
| 47 |
+
- total_train_batch_size: 16
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 1
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 56 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 57 |
+
| 2.1688 | 0.9994 | 1350 | 2.1641 |
|
| 58 |
+
|
| 59 |
+
|
| 60 |
+
### Framework versions
|
| 61 |
+
|
| 62 |
+
- Transformers 4.43.3
|
| 63 |
+
- Pytorch 2.1.2
|
| 64 |
+
- Datasets 2.20.0
|
| 65 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9993522716757657,
|
| 3 |
+
"total_flos": 5.373491620442276e+17,
|
| 4 |
+
"train_loss": 2.19371005181913,
|
| 5 |
+
"train_runtime": 17533.5186,
|
| 6 |
+
"train_samples": 116722,
|
| 7 |
+
"train_samples_per_second": 1.233,
|
| 8 |
+
"train_steps_per_second": 0.077
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 2,
|
| 4 |
+
"cache_implementation": "hybrid",
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
1,
|
| 7 |
+
107
|
| 8 |
+
],
|
| 9 |
+
"pad_token_id": 0,
|
| 10 |
+
"transformers_version": "4.43.3"
|
| 11 |
+
}
|
runs/Aug02_22-17-19_poseidon/events.out.tfevents.1722637052.poseidon.1348942.0
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 sha256:146d5849e2f397291c64aa9cbb9df857601cb39fc4d6999b6921561db41d7a81
|
| 3 |
+
size 63008
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9993522716757657,
|
| 3 |
+
"total_flos": 5.373491620442276e+17,
|
| 4 |
+
"train_loss": 2.19371005181913,
|
| 5 |
+
"train_runtime": 17533.5186,
|
| 6 |
+
"train_samples": 116722,
|
| 7 |
+
"train_samples_per_second": 1.233,
|
| 8 |
+
"train_steps_per_second": 0.077
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1947 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9993522716757657,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1350,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0007402609419820486,
|
| 13 |
+
"grad_norm": 13.5625,
|
| 14 |
+
"learning_rate": 1.4814814814814817e-07,
|
| 15 |
+
"loss": 2.7636,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0037013047099102433,
|
| 20 |
+
"grad_norm": 13.5,
|
| 21 |
+
"learning_rate": 7.407407407407407e-07,
|
| 22 |
+
"loss": 2.7044,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.007402609419820487,
|
| 27 |
+
"grad_norm": 12.0625,
|
| 28 |
+
"learning_rate": 1.4814814814814815e-06,
|
| 29 |
+
"loss": 2.6933,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01110391412973073,
|
| 34 |
+
"grad_norm": 9.75,
|
| 35 |
+
"learning_rate": 2.222222222222222e-06,
|
| 36 |
+
"loss": 2.6397,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.014805218839640973,
|
| 41 |
+
"grad_norm": 9.25,
|
| 42 |
+
"learning_rate": 2.962962962962963e-06,
|
| 43 |
+
"loss": 2.5469,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.018506523549551217,
|
| 48 |
+
"grad_norm": 4.3125,
|
| 49 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 50 |
+
"loss": 2.4799,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02220782825946146,
|
| 55 |
+
"grad_norm": 3.4375,
|
| 56 |
+
"learning_rate": 4.444444444444444e-06,
|
| 57 |
+
"loss": 2.4496,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.025909132969371702,
|
| 62 |
+
"grad_norm": 3.171875,
|
| 63 |
+
"learning_rate": 5.185185185185185e-06,
|
| 64 |
+
"loss": 2.3571,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.029610437679281947,
|
| 69 |
+
"grad_norm": 2.8125,
|
| 70 |
+
"learning_rate": 5.925925925925926e-06,
|
| 71 |
+
"loss": 2.3253,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03331174238919219,
|
| 76 |
+
"grad_norm": 2.75,
|
| 77 |
+
"learning_rate": 6.666666666666667e-06,
|
| 78 |
+
"loss": 2.3086,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.037013047099102435,
|
| 83 |
+
"grad_norm": 2.6875,
|
| 84 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 85 |
+
"loss": 2.2763,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04071435180901268,
|
| 90 |
+
"grad_norm": 2.375,
|
| 91 |
+
"learning_rate": 8.148148148148148e-06,
|
| 92 |
+
"loss": 2.2683,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04441565651892292,
|
| 97 |
+
"grad_norm": 2.515625,
|
| 98 |
+
"learning_rate": 8.888888888888888e-06,
|
| 99 |
+
"loss": 2.2694,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04811696122883316,
|
| 104 |
+
"grad_norm": 2.25,
|
| 105 |
+
"learning_rate": 9.62962962962963e-06,
|
| 106 |
+
"loss": 2.2582,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.051818265938743405,
|
| 111 |
+
"grad_norm": 2.203125,
|
| 112 |
+
"learning_rate": 1.037037037037037e-05,
|
| 113 |
+
"loss": 2.2606,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05551957064865365,
|
| 118 |
+
"grad_norm": 2.1875,
|
| 119 |
+
"learning_rate": 1.1111111111111113e-05,
|
| 120 |
+
"loss": 2.2405,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.05922087535856389,
|
| 125 |
+
"grad_norm": 2.1875,
|
| 126 |
+
"learning_rate": 1.1851851851851852e-05,
|
| 127 |
+
"loss": 2.2496,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06292218006847414,
|
| 132 |
+
"grad_norm": 2.1875,
|
| 133 |
+
"learning_rate": 1.2592592592592593e-05,
|
| 134 |
+
"loss": 2.2547,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06662348477838438,
|
| 139 |
+
"grad_norm": 2.109375,
|
| 140 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 141 |
+
"loss": 2.2488,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07032478948829463,
|
| 146 |
+
"grad_norm": 2.109375,
|
| 147 |
+
"learning_rate": 1.4074074074074075e-05,
|
| 148 |
+
"loss": 2.2479,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07402609419820487,
|
| 153 |
+
"grad_norm": 2.03125,
|
| 154 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 155 |
+
"loss": 2.2405,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.07772739890811511,
|
| 160 |
+
"grad_norm": 2.109375,
|
| 161 |
+
"learning_rate": 1.555555555555556e-05,
|
| 162 |
+
"loss": 2.222,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08142870361802536,
|
| 167 |
+
"grad_norm": 2.140625,
|
| 168 |
+
"learning_rate": 1.6296296296296297e-05,
|
| 169 |
+
"loss": 2.2254,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0851300083279356,
|
| 174 |
+
"grad_norm": 2.0625,
|
| 175 |
+
"learning_rate": 1.7037037037037038e-05,
|
| 176 |
+
"loss": 2.2505,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08883131303784585,
|
| 181 |
+
"grad_norm": 2.015625,
|
| 182 |
+
"learning_rate": 1.7777777777777777e-05,
|
| 183 |
+
"loss": 2.2367,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09253261774775609,
|
| 188 |
+
"grad_norm": 2.1875,
|
| 189 |
+
"learning_rate": 1.851851851851852e-05,
|
| 190 |
+
"loss": 2.2275,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.09623392245766632,
|
| 195 |
+
"grad_norm": 2.078125,
|
| 196 |
+
"learning_rate": 1.925925925925926e-05,
|
| 197 |
+
"loss": 2.2258,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.09993522716757657,
|
| 202 |
+
"grad_norm": 2.03125,
|
| 203 |
+
"learning_rate": 2e-05,
|
| 204 |
+
"loss": 2.2103,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.10363653187748681,
|
| 209 |
+
"grad_norm": 2.015625,
|
| 210 |
+
"learning_rate": 1.9999164298554375e-05,
|
| 211 |
+
"loss": 2.2442,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.10733783658739705,
|
| 216 |
+
"grad_norm": 2.109375,
|
| 217 |
+
"learning_rate": 1.9996657333896875e-05,
|
| 218 |
+
"loss": 2.2542,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1110391412973073,
|
| 223 |
+
"grad_norm": 1.9765625,
|
| 224 |
+
"learning_rate": 1.9992479525042305e-05,
|
| 225 |
+
"loss": 2.2434,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11474044600721754,
|
| 230 |
+
"grad_norm": 2.015625,
|
| 231 |
+
"learning_rate": 1.9986631570270835e-05,
|
| 232 |
+
"loss": 2.2195,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.11844175071712779,
|
| 237 |
+
"grad_norm": 2.03125,
|
| 238 |
+
"learning_rate": 1.9979114447011323e-05,
|
| 239 |
+
"loss": 2.2221,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12214305542703803,
|
| 244 |
+
"grad_norm": 2.0,
|
| 245 |
+
"learning_rate": 1.996992941167792e-05,
|
| 246 |
+
"loss": 2.2276,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12584436013694827,
|
| 251 |
+
"grad_norm": 2.03125,
|
| 252 |
+
"learning_rate": 1.9959077999460094e-05,
|
| 253 |
+
"loss": 2.2112,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1295456648468585,
|
| 258 |
+
"grad_norm": 2.046875,
|
| 259 |
+
"learning_rate": 1.9946562024066018e-05,
|
| 260 |
+
"loss": 2.2086,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13324696955676876,
|
| 265 |
+
"grad_norm": 1.9765625,
|
| 266 |
+
"learning_rate": 1.9932383577419432e-05,
|
| 267 |
+
"loss": 2.2095,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.136948274266679,
|
| 272 |
+
"grad_norm": 1.9921875,
|
| 273 |
+
"learning_rate": 1.991654502931001e-05,
|
| 274 |
+
"loss": 2.2194,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.14064957897658925,
|
| 279 |
+
"grad_norm": 1.953125,
|
| 280 |
+
"learning_rate": 1.9899049026997272e-05,
|
| 281 |
+
"loss": 2.229,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14435088368649948,
|
| 286 |
+
"grad_norm": 1.9921875,
|
| 287 |
+
"learning_rate": 1.9879898494768093e-05,
|
| 288 |
+
"loss": 2.2252,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.14805218839640974,
|
| 293 |
+
"grad_norm": 1.9140625,
|
| 294 |
+
"learning_rate": 1.9859096633447965e-05,
|
| 295 |
+
"loss": 2.2168,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15175349310631997,
|
| 300 |
+
"grad_norm": 2.046875,
|
| 301 |
+
"learning_rate": 1.9836646919866012e-05,
|
| 302 |
+
"loss": 2.2246,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15545479781623023,
|
| 307 |
+
"grad_norm": 1.9375,
|
| 308 |
+
"learning_rate": 1.9812553106273848e-05,
|
| 309 |
+
"loss": 2.2151,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.15915610252614046,
|
| 314 |
+
"grad_norm": 1.90625,
|
| 315 |
+
"learning_rate": 1.9786819219718443e-05,
|
| 316 |
+
"loss": 2.2176,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16285740723605072,
|
| 321 |
+
"grad_norm": 1.9453125,
|
| 322 |
+
"learning_rate": 1.9759449561369036e-05,
|
| 323 |
+
"loss": 2.2279,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16655871194596095,
|
| 328 |
+
"grad_norm": 2.015625,
|
| 329 |
+
"learning_rate": 1.973044870579824e-05,
|
| 330 |
+
"loss": 2.2113,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1702600166558712,
|
| 335 |
+
"grad_norm": 1.875,
|
| 336 |
+
"learning_rate": 1.9699821500217436e-05,
|
| 337 |
+
"loss": 2.2418,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17396132136578144,
|
| 342 |
+
"grad_norm": 1.8984375,
|
| 343 |
+
"learning_rate": 1.9667573063666622e-05,
|
| 344 |
+
"loss": 2.2206,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.1776626260756917,
|
| 349 |
+
"grad_norm": 1.9375,
|
| 350 |
+
"learning_rate": 1.9633708786158803e-05,
|
| 351 |
+
"loss": 2.2162,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.18136393078560192,
|
| 356 |
+
"grad_norm": 1.8828125,
|
| 357 |
+
"learning_rate": 1.959823432777912e-05,
|
| 358 |
+
"loss": 2.2015,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18506523549551218,
|
| 363 |
+
"grad_norm": 1.90625,
|
| 364 |
+
"learning_rate": 1.95611556177388e-05,
|
| 365 |
+
"loss": 2.2095,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.1887665402054224,
|
| 370 |
+
"grad_norm": 1.828125,
|
| 371 |
+
"learning_rate": 1.9522478853384154e-05,
|
| 372 |
+
"loss": 2.2245,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.19246784491533264,
|
| 377 |
+
"grad_norm": 1.8984375,
|
| 378 |
+
"learning_rate": 1.9482210499160767e-05,
|
| 379 |
+
"loss": 2.2179,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.1961691496252429,
|
| 384 |
+
"grad_norm": 1.90625,
|
| 385 |
+
"learning_rate": 1.9440357285533e-05,
|
| 386 |
+
"loss": 2.2132,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.19987045433515313,
|
| 391 |
+
"grad_norm": 2.046875,
|
| 392 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 393 |
+
"loss": 2.2024,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2035717590450634,
|
| 398 |
+
"grad_norm": 1.859375,
|
| 399 |
+
"learning_rate": 1.93519245252219e-05,
|
| 400 |
+
"loss": 2.1749,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.20727306375497362,
|
| 405 |
+
"grad_norm": 2.171875,
|
| 406 |
+
"learning_rate": 1.9305359759215686e-05,
|
| 407 |
+
"loss": 2.2137,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.21097436846488388,
|
| 412 |
+
"grad_norm": 1.921875,
|
| 413 |
+
"learning_rate": 1.9257239692688907e-05,
|
| 414 |
+
"loss": 2.2294,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2146756731747941,
|
| 419 |
+
"grad_norm": 1.90625,
|
| 420 |
+
"learning_rate": 1.9207572368443386e-05,
|
| 421 |
+
"loss": 2.2023,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21837697788470437,
|
| 426 |
+
"grad_norm": 1.890625,
|
| 427 |
+
"learning_rate": 1.9156366087890062e-05,
|
| 428 |
+
"loss": 2.2113,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.2220782825946146,
|
| 433 |
+
"grad_norm": 1.9375,
|
| 434 |
+
"learning_rate": 1.9103629409661468e-05,
|
| 435 |
+
"loss": 2.2245,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22577958730452485,
|
| 440 |
+
"grad_norm": 1.90625,
|
| 441 |
+
"learning_rate": 1.9049371148181253e-05,
|
| 442 |
+
"loss": 2.2102,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.22948089201443508,
|
| 447 |
+
"grad_norm": 1.9609375,
|
| 448 |
+
"learning_rate": 1.8993600372190933e-05,
|
| 449 |
+
"loss": 2.2243,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.23318219672434534,
|
| 454 |
+
"grad_norm": 1.90625,
|
| 455 |
+
"learning_rate": 1.8936326403234125e-05,
|
| 456 |
+
"loss": 2.2009,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.23688350143425557,
|
| 461 |
+
"grad_norm": 1.828125,
|
| 462 |
+
"learning_rate": 1.8877558814098564e-05,
|
| 463 |
+
"loss": 2.2078,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.24058480614416583,
|
| 468 |
+
"grad_norm": 1.8515625,
|
| 469 |
+
"learning_rate": 1.881730742721608e-05,
|
| 470 |
+
"loss": 2.2031,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24428611085407606,
|
| 475 |
+
"grad_norm": 1.8515625,
|
| 476 |
+
"learning_rate": 1.8755582313020912e-05,
|
| 477 |
+
"loss": 2.1597,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.24798741556398632,
|
| 482 |
+
"grad_norm": 1.875,
|
| 483 |
+
"learning_rate": 1.8692393788266477e-05,
|
| 484 |
+
"loss": 2.1922,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.25168872027389655,
|
| 489 |
+
"grad_norm": 1.8984375,
|
| 490 |
+
"learning_rate": 1.8627752414301087e-05,
|
| 491 |
+
"loss": 2.1883,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.2553900249838068,
|
| 496 |
+
"grad_norm": 1.90625,
|
| 497 |
+
"learning_rate": 1.8561668995302668e-05,
|
| 498 |
+
"loss": 2.2097,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.259091329693717,
|
| 503 |
+
"grad_norm": 1.875,
|
| 504 |
+
"learning_rate": 1.8494154576472976e-05,
|
| 505 |
+
"loss": 2.2106,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.26279263440362727,
|
| 510 |
+
"grad_norm": 1.875,
|
| 511 |
+
"learning_rate": 1.8425220442191496e-05,
|
| 512 |
+
"loss": 2.2035,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2664939391135375,
|
| 517 |
+
"grad_norm": 1.859375,
|
| 518 |
+
"learning_rate": 1.8354878114129368e-05,
|
| 519 |
+
"loss": 2.1937,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.2701952438234478,
|
| 524 |
+
"grad_norm": 1.8046875,
|
| 525 |
+
"learning_rate": 1.8283139349323632e-05,
|
| 526 |
+
"loss": 2.1828,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.273896548533358,
|
| 531 |
+
"grad_norm": 1.84375,
|
| 532 |
+
"learning_rate": 1.8210016138212186e-05,
|
| 533 |
+
"loss": 2.189,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.27759785324326824,
|
| 538 |
+
"grad_norm": 1.8203125,
|
| 539 |
+
"learning_rate": 1.8135520702629677e-05,
|
| 540 |
+
"loss": 2.2025,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.2812991579531785,
|
| 545 |
+
"grad_norm": 1.859375,
|
| 546 |
+
"learning_rate": 1.8059665493764745e-05,
|
| 547 |
+
"loss": 2.1967,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.28500046266308876,
|
| 552 |
+
"grad_norm": 1.796875,
|
| 553 |
+
"learning_rate": 1.7982463190078928e-05,
|
| 554 |
+
"loss": 2.1726,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.28870176737299896,
|
| 559 |
+
"grad_norm": 1.8203125,
|
| 560 |
+
"learning_rate": 1.7903926695187595e-05,
|
| 561 |
+
"loss": 2.1758,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.2924030720829092,
|
| 566 |
+
"grad_norm": 1.859375,
|
| 567 |
+
"learning_rate": 1.78240691357032e-05,
|
| 568 |
+
"loss": 2.186,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2961043767928195,
|
| 573 |
+
"grad_norm": 1.828125,
|
| 574 |
+
"learning_rate": 1.7742903859041324e-05,
|
| 575 |
+
"loss": 2.1866,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.29980568150272974,
|
| 580 |
+
"grad_norm": 1.828125,
|
| 581 |
+
"learning_rate": 1.766044443118978e-05,
|
| 582 |
+
"loss": 2.1996,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.30350698621263994,
|
| 587 |
+
"grad_norm": 1.7890625,
|
| 588 |
+
"learning_rate": 1.757670463444118e-05,
|
| 589 |
+
"loss": 2.1657,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3072082909225502,
|
| 594 |
+
"grad_norm": 1.8046875,
|
| 595 |
+
"learning_rate": 1.749169846508936e-05,
|
| 596 |
+
"loss": 2.1938,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.31090959563246046,
|
| 601 |
+
"grad_norm": 1.8359375,
|
| 602 |
+
"learning_rate": 1.740544013109005e-05,
|
| 603 |
+
"loss": 2.1802,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.31461090034237066,
|
| 608 |
+
"grad_norm": 1.8359375,
|
| 609 |
+
"learning_rate": 1.7317944049686125e-05,
|
| 610 |
+
"loss": 2.1961,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3183122050522809,
|
| 615 |
+
"grad_norm": 1.8359375,
|
| 616 |
+
"learning_rate": 1.722922484499793e-05,
|
| 617 |
+
"loss": 2.1849,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.3220135097621912,
|
| 622 |
+
"grad_norm": 1.84375,
|
| 623 |
+
"learning_rate": 1.7139297345578992e-05,
|
| 624 |
+
"loss": 2.2075,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.32571481447210143,
|
| 629 |
+
"grad_norm": 1.8046875,
|
| 630 |
+
"learning_rate": 1.7048176581937562e-05,
|
| 631 |
+
"loss": 2.1531,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.32941611918201164,
|
| 636 |
+
"grad_norm": 1.8359375,
|
| 637 |
+
"learning_rate": 1.6955877784024418e-05,
|
| 638 |
+
"loss": 2.1763,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.3331174238919219,
|
| 643 |
+
"grad_norm": 1.8359375,
|
| 644 |
+
"learning_rate": 1.686241637868734e-05,
|
| 645 |
+
"loss": 2.2074,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.33681872860183215,
|
| 650 |
+
"grad_norm": 1.765625,
|
| 651 |
+
"learning_rate": 1.676780798709262e-05,
|
| 652 |
+
"loss": 2.1793,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3405200333117424,
|
| 657 |
+
"grad_norm": 1.8203125,
|
| 658 |
+
"learning_rate": 1.6672068422114195e-05,
|
| 659 |
+
"loss": 2.1853,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3442213380216526,
|
| 664 |
+
"grad_norm": 1.796875,
|
| 665 |
+
"learning_rate": 1.657521368569064e-05,
|
| 666 |
+
"loss": 2.1804,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.34792264273156287,
|
| 671 |
+
"grad_norm": 1.8515625,
|
| 672 |
+
"learning_rate": 1.647725996615059e-05,
|
| 673 |
+
"loss": 2.1836,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.35162394744147313,
|
| 678 |
+
"grad_norm": 1.9921875,
|
| 679 |
+
"learning_rate": 1.637822363550706e-05,
|
| 680 |
+
"loss": 2.1714,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3553252521513834,
|
| 685 |
+
"grad_norm": 1.953125,
|
| 686 |
+
"learning_rate": 1.627812124672099e-05,
|
| 687 |
+
"loss": 2.2257,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3590265568612936,
|
| 692 |
+
"grad_norm": 1.8515625,
|
| 693 |
+
"learning_rate": 1.6176969530934573e-05,
|
| 694 |
+
"loss": 2.1983,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.36272786157120385,
|
| 699 |
+
"grad_norm": 1.8125,
|
| 700 |
+
"learning_rate": 1.6074785394674835e-05,
|
| 701 |
+
"loss": 2.1925,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3664291662811141,
|
| 706 |
+
"grad_norm": 1.8125,
|
| 707 |
+
"learning_rate": 1.5971585917027864e-05,
|
| 708 |
+
"loss": 2.1606,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.37013047099102436,
|
| 713 |
+
"grad_norm": 1.8125,
|
| 714 |
+
"learning_rate": 1.586738834678418e-05,
|
| 715 |
+
"loss": 2.1738,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.37383177570093457,
|
| 720 |
+
"grad_norm": 1.8203125,
|
| 721 |
+
"learning_rate": 1.5762210099555804e-05,
|
| 722 |
+
"loss": 2.17,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3775330804108448,
|
| 727 |
+
"grad_norm": 1.859375,
|
| 728 |
+
"learning_rate": 1.5656068754865388e-05,
|
| 729 |
+
"loss": 2.1759,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3812343851207551,
|
| 734 |
+
"grad_norm": 1.8359375,
|
| 735 |
+
"learning_rate": 1.554898205320797e-05,
|
| 736 |
+
"loss": 2.2016,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3849356898306653,
|
| 741 |
+
"grad_norm": 1.796875,
|
| 742 |
+
"learning_rate": 1.5440967893085827e-05,
|
| 743 |
+
"loss": 2.1711,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.38863699454057554,
|
| 748 |
+
"grad_norm": 1.84375,
|
| 749 |
+
"learning_rate": 1.5332044328016916e-05,
|
| 750 |
+
"loss": 2.1809,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.3923382992504858,
|
| 755 |
+
"grad_norm": 1.796875,
|
| 756 |
+
"learning_rate": 1.5222229563517385e-05,
|
| 757 |
+
"loss": 2.2018,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.39603960396039606,
|
| 762 |
+
"grad_norm": 1.765625,
|
| 763 |
+
"learning_rate": 1.5111541954058733e-05,
|
| 764 |
+
"loss": 2.1762,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.39974090867030626,
|
| 769 |
+
"grad_norm": 1.7578125,
|
| 770 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 771 |
+
"loss": 2.1955,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4034422133802165,
|
| 776 |
+
"grad_norm": 1.7890625,
|
| 777 |
+
"learning_rate": 1.4887622344495643e-05,
|
| 778 |
+
"loss": 2.1855,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4071435180901268,
|
| 783 |
+
"grad_norm": 1.8203125,
|
| 784 |
+
"learning_rate": 1.4774427770379492e-05,
|
| 785 |
+
"loss": 2.174,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.41084482280003704,
|
| 790 |
+
"grad_norm": 1.8125,
|
| 791 |
+
"learning_rate": 1.4660435197025391e-05,
|
| 792 |
+
"loss": 2.1727,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.41454612750994724,
|
| 797 |
+
"grad_norm": 1.828125,
|
| 798 |
+
"learning_rate": 1.4545663677185007e-05,
|
| 799 |
+
"loss": 2.1715,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4182474322198575,
|
| 804 |
+
"grad_norm": 1.8203125,
|
| 805 |
+
"learning_rate": 1.4430132393803353e-05,
|
| 806 |
+
"loss": 2.1893,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.42194873692976775,
|
| 811 |
+
"grad_norm": 1.8046875,
|
| 812 |
+
"learning_rate": 1.4313860656812537e-05,
|
| 813 |
+
"loss": 2.1734,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.425650041639678,
|
| 818 |
+
"grad_norm": 1.8046875,
|
| 819 |
+
"learning_rate": 1.4196867899904292e-05,
|
| 820 |
+
"loss": 2.1759,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4293513463495882,
|
| 825 |
+
"grad_norm": 1.78125,
|
| 826 |
+
"learning_rate": 1.4079173677281836e-05,
|
| 827 |
+
"loss": 2.1615,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4330526510594985,
|
| 832 |
+
"grad_norm": 1.796875,
|
| 833 |
+
"learning_rate": 1.396079766039157e-05,
|
| 834 |
+
"loss": 2.1769,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.43675395576940873,
|
| 839 |
+
"grad_norm": 1.8125,
|
| 840 |
+
"learning_rate": 1.3841759634635177e-05,
|
| 841 |
+
"loss": 2.1867,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.44045526047931893,
|
| 846 |
+
"grad_norm": 1.8046875,
|
| 847 |
+
"learning_rate": 1.3722079496062702e-05,
|
| 848 |
+
"loss": 2.1836,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.4441565651892292,
|
| 853 |
+
"grad_norm": 1.765625,
|
| 854 |
+
"learning_rate": 1.3601777248047105e-05,
|
| 855 |
+
"loss": 2.1803,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.44785786989913945,
|
| 860 |
+
"grad_norm": 1.828125,
|
| 861 |
+
"learning_rate": 1.3480872997940906e-05,
|
| 862 |
+
"loss": 2.1667,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4515591746090497,
|
| 867 |
+
"grad_norm": 1.8203125,
|
| 868 |
+
"learning_rate": 1.3359386953715423e-05,
|
| 869 |
+
"loss": 2.1644,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4552604793189599,
|
| 874 |
+
"grad_norm": 1.8125,
|
| 875 |
+
"learning_rate": 1.3237339420583213e-05,
|
| 876 |
+
"loss": 2.1769,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.45896178402887017,
|
| 881 |
+
"grad_norm": 1.78125,
|
| 882 |
+
"learning_rate": 1.3114750797604248e-05,
|
| 883 |
+
"loss": 2.1611,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.4626630887387804,
|
| 888 |
+
"grad_norm": 1.796875,
|
| 889 |
+
"learning_rate": 1.2991641574276419e-05,
|
| 890 |
+
"loss": 2.1676,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.4663643934486907,
|
| 895 |
+
"grad_norm": 1.8671875,
|
| 896 |
+
"learning_rate": 1.2868032327110904e-05,
|
| 897 |
+
"loss": 2.2038,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.4700656981586009,
|
| 902 |
+
"grad_norm": 1.8203125,
|
| 903 |
+
"learning_rate": 1.2743943716193017e-05,
|
| 904 |
+
"loss": 2.2025,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.47376700286851114,
|
| 909 |
+
"grad_norm": 1.7890625,
|
| 910 |
+
"learning_rate": 1.261939648172906e-05,
|
| 911 |
+
"loss": 2.1784,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.4774683075784214,
|
| 916 |
+
"grad_norm": 1.796875,
|
| 917 |
+
"learning_rate": 1.2494411440579814e-05,
|
| 918 |
+
"loss": 2.1805,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.48116961228833166,
|
| 923 |
+
"grad_norm": 1.796875,
|
| 924 |
+
"learning_rate": 1.2369009482781191e-05,
|
| 925 |
+
"loss": 2.1945,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.48487091699824186,
|
| 930 |
+
"grad_norm": 1.8359375,
|
| 931 |
+
"learning_rate": 1.2243211568052678e-05,
|
| 932 |
+
"loss": 2.1775,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.4885722217081521,
|
| 937 |
+
"grad_norm": 1.796875,
|
| 938 |
+
"learning_rate": 1.211703872229411e-05,
|
| 939 |
+
"loss": 2.1832,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4922735264180624,
|
| 944 |
+
"grad_norm": 1.84375,
|
| 945 |
+
"learning_rate": 1.1990512034071407e-05,
|
| 946 |
+
"loss": 2.1899,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.49597483112797264,
|
| 951 |
+
"grad_norm": 1.8203125,
|
| 952 |
+
"learning_rate": 1.1863652651091824e-05,
|
| 953 |
+
"loss": 2.1675,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.49967613583788284,
|
| 958 |
+
"grad_norm": 1.7734375,
|
| 959 |
+
"learning_rate": 1.1736481776669307e-05,
|
| 960 |
+
"loss": 2.1881,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5033774405477931,
|
| 965 |
+
"grad_norm": 1.7890625,
|
| 966 |
+
"learning_rate": 1.1609020666180574e-05,
|
| 967 |
+
"loss": 2.1825,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5070787452577034,
|
| 972 |
+
"grad_norm": 1.78125,
|
| 973 |
+
"learning_rate": 1.1481290623512491e-05,
|
| 974 |
+
"loss": 2.1875,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5107800499676136,
|
| 979 |
+
"grad_norm": 1.828125,
|
| 980 |
+
"learning_rate": 1.1353312997501313e-05,
|
| 981 |
+
"loss": 2.1693,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5144813546775239,
|
| 986 |
+
"grad_norm": 1.78125,
|
| 987 |
+
"learning_rate": 1.1225109178364456e-05,
|
| 988 |
+
"loss": 2.169,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.518182659387434,
|
| 993 |
+
"grad_norm": 1.765625,
|
| 994 |
+
"learning_rate": 1.1096700594125318e-05,
|
| 995 |
+
"loss": 2.1714,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.5218839640973443,
|
| 1000 |
+
"grad_norm": 1.8203125,
|
| 1001 |
+
"learning_rate": 1.0968108707031792e-05,
|
| 1002 |
+
"loss": 2.179,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5255852688072545,
|
| 1007 |
+
"grad_norm": 1.765625,
|
| 1008 |
+
"learning_rate": 1.0839355009969068e-05,
|
| 1009 |
+
"loss": 2.153,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5292865735171648,
|
| 1014 |
+
"grad_norm": 1.796875,
|
| 1015 |
+
"learning_rate": 1.0710461022867303e-05,
|
| 1016 |
+
"loss": 2.1683,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.532987878227075,
|
| 1021 |
+
"grad_norm": 1.7578125,
|
| 1022 |
+
"learning_rate": 1.0581448289104759e-05,
|
| 1023 |
+
"loss": 2.1459,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.5366891829369853,
|
| 1028 |
+
"grad_norm": 1.8515625,
|
| 1029 |
+
"learning_rate": 1.0452338371907065e-05,
|
| 1030 |
+
"loss": 2.1666,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.5403904876468956,
|
| 1035 |
+
"grad_norm": 1.765625,
|
| 1036 |
+
"learning_rate": 1.0323152850743107e-05,
|
| 1037 |
+
"loss": 2.173,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5440917923568058,
|
| 1042 |
+
"grad_norm": 1.8046875,
|
| 1043 |
+
"learning_rate": 1.0193913317718245e-05,
|
| 1044 |
+
"loss": 2.1745,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.547793097066716,
|
| 1049 |
+
"grad_norm": 1.7734375,
|
| 1050 |
+
"learning_rate": 1.0064641373965394e-05,
|
| 1051 |
+
"loss": 2.1602,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5514944017766262,
|
| 1056 |
+
"grad_norm": 1.7578125,
|
| 1057 |
+
"learning_rate": 9.935358626034607e-06,
|
| 1058 |
+
"loss": 2.1608,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5551957064865365,
|
| 1063 |
+
"grad_norm": 1.796875,
|
| 1064 |
+
"learning_rate": 9.806086682281759e-06,
|
| 1065 |
+
"loss": 2.1906,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5588970111964467,
|
| 1070 |
+
"grad_norm": 1.7734375,
|
| 1071 |
+
"learning_rate": 9.676847149256894e-06,
|
| 1072 |
+
"loss": 2.1502,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.562598315906357,
|
| 1077 |
+
"grad_norm": 1.765625,
|
| 1078 |
+
"learning_rate": 9.547661628092938e-06,
|
| 1079 |
+
"loss": 2.1648,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5662996206162673,
|
| 1084 |
+
"grad_norm": 1.8515625,
|
| 1085 |
+
"learning_rate": 9.418551710895243e-06,
|
| 1086 |
+
"loss": 2.1866,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5700009253261775,
|
| 1091 |
+
"grad_norm": 1.7734375,
|
| 1092 |
+
"learning_rate": 9.289538977132702e-06,
|
| 1093 |
+
"loss": 2.1776,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5737022300360877,
|
| 1098 |
+
"grad_norm": 1.7734375,
|
| 1099 |
+
"learning_rate": 9.160644990030932e-06,
|
| 1100 |
+
"loss": 2.1665,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5774035347459979,
|
| 1105 |
+
"grad_norm": 1.7734375,
|
| 1106 |
+
"learning_rate": 9.03189129296821e-06,
|
| 1107 |
+
"loss": 2.1891,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5811048394559082,
|
| 1112 |
+
"grad_norm": 1.8203125,
|
| 1113 |
+
"learning_rate": 8.903299405874685e-06,
|
| 1114 |
+
"loss": 2.1894,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5848061441658184,
|
| 1119 |
+
"grad_norm": 1.7734375,
|
| 1120 |
+
"learning_rate": 8.774890821635548e-06,
|
| 1121 |
+
"loss": 2.167,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5885074488757287,
|
| 1126 |
+
"grad_norm": 1.8359375,
|
| 1127 |
+
"learning_rate": 8.646687002498692e-06,
|
| 1128 |
+
"loss": 2.1759,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.592208753585639,
|
| 1133 |
+
"grad_norm": 1.8046875,
|
| 1134 |
+
"learning_rate": 8.518709376487515e-06,
|
| 1135 |
+
"loss": 2.155,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5959100582955492,
|
| 1140 |
+
"grad_norm": 1.796875,
|
| 1141 |
+
"learning_rate": 8.390979333819427e-06,
|
| 1142 |
+
"loss": 2.1734,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5996113630054595,
|
| 1147 |
+
"grad_norm": 1.7890625,
|
| 1148 |
+
"learning_rate": 8.263518223330698e-06,
|
| 1149 |
+
"loss": 2.1619,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.6033126677153696,
|
| 1154 |
+
"grad_norm": 1.8125,
|
| 1155 |
+
"learning_rate": 8.13634734890818e-06,
|
| 1156 |
+
"loss": 2.1509,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6070139724252799,
|
| 1161 |
+
"grad_norm": 1.796875,
|
| 1162 |
+
"learning_rate": 8.009487965928597e-06,
|
| 1163 |
+
"loss": 2.1662,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6107152771351901,
|
| 1168 |
+
"grad_norm": 1.796875,
|
| 1169 |
+
"learning_rate": 7.882961277705897e-06,
|
| 1170 |
+
"loss": 2.1315,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6144165818451004,
|
| 1175 |
+
"grad_norm": 1.8046875,
|
| 1176 |
+
"learning_rate": 7.756788431947327e-06,
|
| 1177 |
+
"loss": 2.1737,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.6181178865550107,
|
| 1182 |
+
"grad_norm": 1.8359375,
|
| 1183 |
+
"learning_rate": 7.630990517218809e-06,
|
| 1184 |
+
"loss": 2.1673,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.6218191912649209,
|
| 1189 |
+
"grad_norm": 1.859375,
|
| 1190 |
+
"learning_rate": 7.505588559420188e-06,
|
| 1191 |
+
"loss": 2.156,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.6255204959748312,
|
| 1196 |
+
"grad_norm": 1.796875,
|
| 1197 |
+
"learning_rate": 7.380603518270942e-06,
|
| 1198 |
+
"loss": 2.1576,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.6292218006847413,
|
| 1203 |
+
"grad_norm": 1.7890625,
|
| 1204 |
+
"learning_rate": 7.256056283806987e-06,
|
| 1205 |
+
"loss": 2.1715,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.6329231053946516,
|
| 1210 |
+
"grad_norm": 1.78125,
|
| 1211 |
+
"learning_rate": 7.131967672889101e-06,
|
| 1212 |
+
"loss": 2.1872,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.6366244101045618,
|
| 1217 |
+
"grad_norm": 1.828125,
|
| 1218 |
+
"learning_rate": 7.008358425723586e-06,
|
| 1219 |
+
"loss": 2.1596,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6403257148144721,
|
| 1224 |
+
"grad_norm": 1.765625,
|
| 1225 |
+
"learning_rate": 6.885249202395754e-06,
|
| 1226 |
+
"loss": 2.1546,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.6440270195243823,
|
| 1231 |
+
"grad_norm": 1.796875,
|
| 1232 |
+
"learning_rate": 6.762660579416791e-06,
|
| 1233 |
+
"loss": 2.1499,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.6477283242342926,
|
| 1238 |
+
"grad_norm": 1.8203125,
|
| 1239 |
+
"learning_rate": 6.640613046284581e-06,
|
| 1240 |
+
"loss": 2.1816,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6514296289442029,
|
| 1245 |
+
"grad_norm": 1.8359375,
|
| 1246 |
+
"learning_rate": 6.519127002059096e-06,
|
| 1247 |
+
"loss": 2.1675,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6551309336541131,
|
| 1252 |
+
"grad_norm": 1.796875,
|
| 1253 |
+
"learning_rate": 6.3982227519528986e-06,
|
| 1254 |
+
"loss": 2.1819,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6588322383640233,
|
| 1259 |
+
"grad_norm": 1.8359375,
|
| 1260 |
+
"learning_rate": 6.277920503937303e-06,
|
| 1261 |
+
"loss": 2.1628,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6625335430739335,
|
| 1266 |
+
"grad_norm": 1.8046875,
|
| 1267 |
+
"learning_rate": 6.158240365364823e-06,
|
| 1268 |
+
"loss": 2.1595,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6662348477838438,
|
| 1273 |
+
"grad_norm": 1.7890625,
|
| 1274 |
+
"learning_rate": 6.039202339608432e-06,
|
| 1275 |
+
"loss": 2.1604,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.669936152493754,
|
| 1280 |
+
"grad_norm": 1.7890625,
|
| 1281 |
+
"learning_rate": 5.920826322718165e-06,
|
| 1282 |
+
"loss": 2.1498,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.6736374572036643,
|
| 1287 |
+
"grad_norm": 1.8125,
|
| 1288 |
+
"learning_rate": 5.80313210009571e-06,
|
| 1289 |
+
"loss": 2.1612,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6773387619135746,
|
| 1294 |
+
"grad_norm": 1.7890625,
|
| 1295 |
+
"learning_rate": 5.686139343187468e-06,
|
| 1296 |
+
"loss": 2.1668,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6810400666234848,
|
| 1301 |
+
"grad_norm": 1.75,
|
| 1302 |
+
"learning_rate": 5.569867606196652e-06,
|
| 1303 |
+
"loss": 2.1952,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6847413713333951,
|
| 1308 |
+
"grad_norm": 1.7578125,
|
| 1309 |
+
"learning_rate": 5.454336322814995e-06,
|
| 1310 |
+
"loss": 2.1699,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6884426760433052,
|
| 1315 |
+
"grad_norm": 1.7578125,
|
| 1316 |
+
"learning_rate": 5.339564802974615e-06,
|
| 1317 |
+
"loss": 2.176,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6921439807532155,
|
| 1322 |
+
"grad_norm": 1.8515625,
|
| 1323 |
+
"learning_rate": 5.2255722296205104e-06,
|
| 1324 |
+
"loss": 2.1927,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6958452854631257,
|
| 1329 |
+
"grad_norm": 1.796875,
|
| 1330 |
+
"learning_rate": 5.112377655504359e-06,
|
| 1331 |
+
"loss": 2.1742,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.699546590173036,
|
| 1336 |
+
"grad_norm": 1.78125,
|
| 1337 |
+
"learning_rate": 5.000000000000003e-06,
|
| 1338 |
+
"loss": 2.1835,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7032478948829463,
|
| 1343 |
+
"grad_norm": 1.84375,
|
| 1344 |
+
"learning_rate": 4.888458045941269e-06,
|
| 1345 |
+
"loss": 2.1855,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7069491995928565,
|
| 1350 |
+
"grad_norm": 1.8125,
|
| 1351 |
+
"learning_rate": 4.7777704364826175e-06,
|
| 1352 |
+
"loss": 2.1922,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7106505043027668,
|
| 1357 |
+
"grad_norm": 1.8046875,
|
| 1358 |
+
"learning_rate": 4.66795567198309e-06,
|
| 1359 |
+
"loss": 2.1563,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.7143518090126769,
|
| 1364 |
+
"grad_norm": 1.78125,
|
| 1365 |
+
"learning_rate": 4.559032106914173e-06,
|
| 1366 |
+
"loss": 2.1522,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.7180531137225872,
|
| 1371 |
+
"grad_norm": 1.8203125,
|
| 1372 |
+
"learning_rate": 4.4510179467920325e-06,
|
| 1373 |
+
"loss": 2.175,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.7217544184324974,
|
| 1378 |
+
"grad_norm": 1.765625,
|
| 1379 |
+
"learning_rate": 4.343931245134616e-06,
|
| 1380 |
+
"loss": 2.1506,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.7254557231424077,
|
| 1385 |
+
"grad_norm": 1.8046875,
|
| 1386 |
+
"learning_rate": 4.237789900444197e-06,
|
| 1387 |
+
"loss": 2.1555,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.729157027852318,
|
| 1392 |
+
"grad_norm": 1.78125,
|
| 1393 |
+
"learning_rate": 4.132611653215822e-06,
|
| 1394 |
+
"loss": 2.1142,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7328583325622282,
|
| 1399 |
+
"grad_norm": 1.7890625,
|
| 1400 |
+
"learning_rate": 4.028414082972141e-06,
|
| 1401 |
+
"loss": 2.1617,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7365596372721385,
|
| 1406 |
+
"grad_norm": 1.8046875,
|
| 1407 |
+
"learning_rate": 3.925214605325164e-06,
|
| 1408 |
+
"loss": 2.1854,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7402609419820487,
|
| 1413 |
+
"grad_norm": 1.7734375,
|
| 1414 |
+
"learning_rate": 3.823030469065431e-06,
|
| 1415 |
+
"loss": 2.1767,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.7439622466919589,
|
| 1420 |
+
"grad_norm": 1.8125,
|
| 1421 |
+
"learning_rate": 3.7218787532790167e-06,
|
| 1422 |
+
"loss": 2.1667,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.7476635514018691,
|
| 1427 |
+
"grad_norm": 1.8046875,
|
| 1428 |
+
"learning_rate": 3.6217763644929393e-06,
|
| 1429 |
+
"loss": 2.1532,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.7513648561117794,
|
| 1434 |
+
"grad_norm": 1.8203125,
|
| 1435 |
+
"learning_rate": 3.522740033849411e-06,
|
| 1436 |
+
"loss": 2.1617,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.7550661608216896,
|
| 1441 |
+
"grad_norm": 1.8828125,
|
| 1442 |
+
"learning_rate": 3.424786314309365e-06,
|
| 1443 |
+
"loss": 2.1549,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7587674655315999,
|
| 1448 |
+
"grad_norm": 1.7265625,
|
| 1449 |
+
"learning_rate": 3.3279315778858034e-06,
|
| 1450 |
+
"loss": 2.1501,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.7624687702415102,
|
| 1455 |
+
"grad_norm": 1.7734375,
|
| 1456 |
+
"learning_rate": 3.2321920129073815e-06,
|
| 1457 |
+
"loss": 2.164,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.7661700749514204,
|
| 1462 |
+
"grad_norm": 1.765625,
|
| 1463 |
+
"learning_rate": 3.1375836213126653e-06,
|
| 1464 |
+
"loss": 2.1473,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7698713796613306,
|
| 1469 |
+
"grad_norm": 1.7890625,
|
| 1470 |
+
"learning_rate": 3.04412221597558e-06,
|
| 1471 |
+
"loss": 2.1597,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7735726843712408,
|
| 1476 |
+
"grad_norm": 1.8359375,
|
| 1477 |
+
"learning_rate": 2.9518234180624393e-06,
|
| 1478 |
+
"loss": 2.165,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7772739890811511,
|
| 1483 |
+
"grad_norm": 1.7890625,
|
| 1484 |
+
"learning_rate": 2.8607026544210115e-06,
|
| 1485 |
+
"loss": 2.1634,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7809752937910613,
|
| 1490 |
+
"grad_norm": 1.7421875,
|
| 1491 |
+
"learning_rate": 2.770775155002071e-06,
|
| 1492 |
+
"loss": 2.1559,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7846765985009716,
|
| 1497 |
+
"grad_norm": 1.7734375,
|
| 1498 |
+
"learning_rate": 2.6820559503138797e-06,
|
| 1499 |
+
"loss": 2.1638,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7883779032108819,
|
| 1504 |
+
"grad_norm": 1.796875,
|
| 1505 |
+
"learning_rate": 2.594559868909956e-06,
|
| 1506 |
+
"loss": 2.1604,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.7920792079207921,
|
| 1511 |
+
"grad_norm": 1.7578125,
|
| 1512 |
+
"learning_rate": 2.50830153491064e-06,
|
| 1513 |
+
"loss": 2.1602,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7957805126307024,
|
| 1518 |
+
"grad_norm": 1.765625,
|
| 1519 |
+
"learning_rate": 2.423295365558821e-06,
|
| 1520 |
+
"loss": 2.1588,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.7994818173406125,
|
| 1525 |
+
"grad_norm": 1.8125,
|
| 1526 |
+
"learning_rate": 2.339555568810221e-06,
|
| 1527 |
+
"loss": 2.1525,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.8031831220505228,
|
| 1532 |
+
"grad_norm": 1.7578125,
|
| 1533 |
+
"learning_rate": 2.2570961409586756e-06,
|
| 1534 |
+
"loss": 2.1622,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.806884426760433,
|
| 1539 |
+
"grad_norm": 1.7890625,
|
| 1540 |
+
"learning_rate": 2.1759308642968024e-06,
|
| 1541 |
+
"loss": 2.1544,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.8105857314703433,
|
| 1546 |
+
"grad_norm": 1.8046875,
|
| 1547 |
+
"learning_rate": 2.0960733048124082e-06,
|
| 1548 |
+
"loss": 2.158,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.8142870361802536,
|
| 1553 |
+
"grad_norm": 1.828125,
|
| 1554 |
+
"learning_rate": 2.01753680992107e-06,
|
| 1555 |
+
"loss": 2.1827,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.8179883408901638,
|
| 1560 |
+
"grad_norm": 1.7890625,
|
| 1561 |
+
"learning_rate": 1.9403345062352574e-06,
|
| 1562 |
+
"loss": 2.1579,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.8216896456000741,
|
| 1567 |
+
"grad_norm": 1.7890625,
|
| 1568 |
+
"learning_rate": 1.8644792973703252e-06,
|
| 1569 |
+
"loss": 2.1601,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.8253909503099842,
|
| 1574 |
+
"grad_norm": 1.8515625,
|
| 1575 |
+
"learning_rate": 1.7899838617878163e-06,
|
| 1576 |
+
"loss": 2.168,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.8290922550198945,
|
| 1581 |
+
"grad_norm": 1.8359375,
|
| 1582 |
+
"learning_rate": 1.7168606506763696e-06,
|
| 1583 |
+
"loss": 2.1706,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.8327935597298047,
|
| 1588 |
+
"grad_norm": 1.796875,
|
| 1589 |
+
"learning_rate": 1.6451218858706374e-06,
|
| 1590 |
+
"loss": 2.175,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.836494864439715,
|
| 1595 |
+
"grad_norm": 1.765625,
|
| 1596 |
+
"learning_rate": 1.5747795578085046e-06,
|
| 1597 |
+
"loss": 2.1803,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.8401961691496252,
|
| 1602 |
+
"grad_norm": 1.7578125,
|
| 1603 |
+
"learning_rate": 1.505845423527027e-06,
|
| 1604 |
+
"loss": 2.1543,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.8438974738595355,
|
| 1609 |
+
"grad_norm": 1.7890625,
|
| 1610 |
+
"learning_rate": 1.4383310046973365e-06,
|
| 1611 |
+
"loss": 2.1556,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.8475987785694458,
|
| 1616 |
+
"grad_norm": 1.828125,
|
| 1617 |
+
"learning_rate": 1.372247585698916e-06,
|
| 1618 |
+
"loss": 2.1772,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.851300083279356,
|
| 1623 |
+
"grad_norm": 1.7734375,
|
| 1624 |
+
"learning_rate": 1.307606211733522e-06,
|
| 1625 |
+
"loss": 2.1542,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.8550013879892662,
|
| 1630 |
+
"grad_norm": 1.8125,
|
| 1631 |
+
"learning_rate": 1.2444176869790925e-06,
|
| 1632 |
+
"loss": 2.1635,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.8587026926991764,
|
| 1637 |
+
"grad_norm": 1.8046875,
|
| 1638 |
+
"learning_rate": 1.18269257278392e-06,
|
| 1639 |
+
"loss": 2.144,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.8624039974090867,
|
| 1644 |
+
"grad_norm": 1.765625,
|
| 1645 |
+
"learning_rate": 1.1224411859014417e-06,
|
| 1646 |
+
"loss": 2.153,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.866105302118997,
|
| 1651 |
+
"grad_norm": 1.84375,
|
| 1652 |
+
"learning_rate": 1.0636735967658785e-06,
|
| 1653 |
+
"loss": 2.1537,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8698066068289072,
|
| 1658 |
+
"grad_norm": 1.7890625,
|
| 1659 |
+
"learning_rate": 1.0063996278090704e-06,
|
| 1660 |
+
"loss": 2.1913,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8735079115388175,
|
| 1665 |
+
"grad_norm": 1.8203125,
|
| 1666 |
+
"learning_rate": 9.506288518187468e-07,
|
| 1667 |
+
"loss": 2.1783,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8772092162487277,
|
| 1672 |
+
"grad_norm": 1.75,
|
| 1673 |
+
"learning_rate": 8.963705903385344e-07,
|
| 1674 |
+
"loss": 2.1502,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8809105209586379,
|
| 1679 |
+
"grad_norm": 1.7890625,
|
| 1680 |
+
"learning_rate": 8.436339121099413e-07,
|
| 1681 |
+
"loss": 2.1642,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.8846118256685481,
|
| 1686 |
+
"grad_norm": 1.7734375,
|
| 1687 |
+
"learning_rate": 7.924276315566171e-07,
|
| 1688 |
+
"loss": 2.1625,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8883131303784584,
|
| 1693 |
+
"grad_norm": 1.8046875,
|
| 1694 |
+
"learning_rate": 7.427603073110967e-07,
|
| 1695 |
+
"loss": 2.1741,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8920144350883686,
|
| 1700 |
+
"grad_norm": 1.796875,
|
| 1701 |
+
"learning_rate": 6.946402407843156e-07,
|
| 1702 |
+
"loss": 2.1467,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8957157397982789,
|
| 1707 |
+
"grad_norm": 1.7578125,
|
| 1708 |
+
"learning_rate": 6.480754747781037e-07,
|
| 1709 |
+
"loss": 2.1597,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8994170445081892,
|
| 1714 |
+
"grad_norm": 1.8046875,
|
| 1715 |
+
"learning_rate": 6.030737921409169e-07,
|
| 1716 |
+
"loss": 2.1653,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.9031183492180994,
|
| 1721 |
+
"grad_norm": 1.796875,
|
| 1722 |
+
"learning_rate": 5.596427144670002e-07,
|
| 1723 |
+
"loss": 2.173,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.9068196539280097,
|
| 1728 |
+
"grad_norm": 1.8125,
|
| 1729 |
+
"learning_rate": 5.177895008392353e-07,
|
| 1730 |
+
"loss": 2.1844,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.9105209586379198,
|
| 1735 |
+
"grad_norm": 1.7109375,
|
| 1736 |
+
"learning_rate": 4.775211466158469e-07,
|
| 1737 |
+
"loss": 2.1385,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.9142222633478301,
|
| 1742 |
+
"grad_norm": 1.7734375,
|
| 1743 |
+
"learning_rate": 4.388443822612043e-07,
|
| 1744 |
+
"loss": 2.1521,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.9179235680577403,
|
| 1749 |
+
"grad_norm": 1.7890625,
|
| 1750 |
+
"learning_rate": 4.017656722208807e-07,
|
| 1751 |
+
"loss": 2.1692,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.9216248727676506,
|
| 1756 |
+
"grad_norm": 1.7421875,
|
| 1757 |
+
"learning_rate": 3.662912138411967e-07,
|
| 1758 |
+
"loss": 2.1767,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.9253261774775609,
|
| 1763 |
+
"grad_norm": 1.796875,
|
| 1764 |
+
"learning_rate": 3.3242693633337986e-07,
|
| 1765 |
+
"loss": 2.1517,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.9290274821874711,
|
| 1770 |
+
"grad_norm": 1.796875,
|
| 1771 |
+
"learning_rate": 3.001784997825652e-07,
|
| 1772 |
+
"loss": 2.1849,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.9327287868973814,
|
| 1777 |
+
"grad_norm": 1.796875,
|
| 1778 |
+
"learning_rate": 2.6955129420176193e-07,
|
| 1779 |
+
"loss": 2.1564,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.9364300916072915,
|
| 1784 |
+
"grad_norm": 1.8125,
|
| 1785 |
+
"learning_rate": 2.405504386309643e-07,
|
| 1786 |
+
"loss": 2.1831,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.9401313963172018,
|
| 1791 |
+
"grad_norm": 1.7734375,
|
| 1792 |
+
"learning_rate": 2.1318078028155886e-07,
|
| 1793 |
+
"loss": 2.1436,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.943832701027112,
|
| 1798 |
+
"grad_norm": 1.796875,
|
| 1799 |
+
"learning_rate": 1.874468937261531e-07,
|
| 1800 |
+
"loss": 2.159,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.9475340057370223,
|
| 1805 |
+
"grad_norm": 1.78125,
|
| 1806 |
+
"learning_rate": 1.6335308013398888e-07,
|
| 1807 |
+
"loss": 2.1677,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.9512353104469325,
|
| 1812 |
+
"grad_norm": 1.7890625,
|
| 1813 |
+
"learning_rate": 1.409033665520354e-07,
|
| 1814 |
+
"loss": 2.1872,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.9549366151568428,
|
| 1819 |
+
"grad_norm": 1.78125,
|
| 1820 |
+
"learning_rate": 1.201015052319099e-07,
|
| 1821 |
+
"loss": 2.1408,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.9586379198667531,
|
| 1826 |
+
"grad_norm": 1.75,
|
| 1827 |
+
"learning_rate": 1.0095097300273026e-07,
|
| 1828 |
+
"loss": 2.1421,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.9623392245766633,
|
| 1833 |
+
"grad_norm": 1.7890625,
|
| 1834 |
+
"learning_rate": 8.345497068998897e-08,
|
| 1835 |
+
"loss": 2.1545,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.9660405292865735,
|
| 1840 |
+
"grad_norm": 1.7734375,
|
| 1841 |
+
"learning_rate": 6.761642258056977e-08,
|
| 1842 |
+
"loss": 2.1459,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.9697418339964837,
|
| 1847 |
+
"grad_norm": 1.7421875,
|
| 1848 |
+
"learning_rate": 5.3437975933985366e-08,
|
| 1849 |
+
"loss": 2.1508,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.973443138706394,
|
| 1854 |
+
"grad_norm": 1.78125,
|
| 1855 |
+
"learning_rate": 4.0922000539906914e-08,
|
| 1856 |
+
"loss": 2.1533,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9771444434163042,
|
| 1861 |
+
"grad_norm": 1.78125,
|
| 1862 |
+
"learning_rate": 3.0070588322079765e-08,
|
| 1863 |
+
"loss": 2.1527,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.9808457481262145,
|
| 1868 |
+
"grad_norm": 1.921875,
|
| 1869 |
+
"learning_rate": 2.088555298867978e-08,
|
| 1870 |
+
"loss": 2.1634,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9845470528361248,
|
| 1875 |
+
"grad_norm": 1.75,
|
| 1876 |
+
"learning_rate": 1.3368429729168075e-08,
|
| 1877 |
+
"loss": 2.1714,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.988248357546035,
|
| 1882 |
+
"grad_norm": 1.75,
|
| 1883 |
+
"learning_rate": 7.520474957699586e-09,
|
| 1884 |
+
"loss": 2.1449,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9919496622559453,
|
| 1889 |
+
"grad_norm": 1.7578125,
|
| 1890 |
+
"learning_rate": 3.3426661031255024e-09,
|
| 1891 |
+
"loss": 2.1756,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9956509669658554,
|
| 1896 |
+
"grad_norm": 1.765625,
|
| 1897 |
+
"learning_rate": 8.357014456272794e-10,
|
| 1898 |
+
"loss": 2.1534,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9993522716757657,
|
| 1903 |
+
"grad_norm": 1.796875,
|
| 1904 |
+
"learning_rate": 0.0,
|
| 1905 |
+
"loss": 2.1688,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.9993522716757657,
|
| 1910 |
+
"eval_loss": 2.16412353515625,
|
| 1911 |
+
"eval_runtime": 221.9604,
|
| 1912 |
+
"eval_samples_per_second": 5.379,
|
| 1913 |
+
"eval_steps_per_second": 2.69,
|
| 1914 |
+
"step": 1350
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 0.9993522716757657,
|
| 1918 |
+
"step": 1350,
|
| 1919 |
+
"total_flos": 5.373491620442276e+17,
|
| 1920 |
+
"train_loss": 2.19371005181913,
|
| 1921 |
+
"train_runtime": 17533.5186,
|
| 1922 |
+
"train_samples_per_second": 1.233,
|
| 1923 |
+
"train_steps_per_second": 0.077
|
| 1924 |
+
}
|
| 1925 |
+
],
|
| 1926 |
+
"logging_steps": 5,
|
| 1927 |
+
"max_steps": 1350,
|
| 1928 |
+
"num_input_tokens_seen": 0,
|
| 1929 |
+
"num_train_epochs": 1,
|
| 1930 |
+
"save_steps": 100,
|
| 1931 |
+
"stateful_callbacks": {
|
| 1932 |
+
"TrainerControl": {
|
| 1933 |
+
"args": {
|
| 1934 |
+
"should_epoch_stop": false,
|
| 1935 |
+
"should_evaluate": false,
|
| 1936 |
+
"should_log": false,
|
| 1937 |
+
"should_save": true,
|
| 1938 |
+
"should_training_stop": true
|
| 1939 |
+
},
|
| 1940 |
+
"attributes": {}
|
| 1941 |
+
}
|
| 1942 |
+
},
|
| 1943 |
+
"total_flos": 5.373491620442276e+17,
|
| 1944 |
+
"train_batch_size": 1,
|
| 1945 |
+
"trial_name": null,
|
| 1946 |
+
"trial_params": null
|
| 1947 |
+
}
|