Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Sep09_09-31-09_COE-CS-sv003/events.out.tfevents.1725874574.COE-CS-sv003.742544.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1359 -0
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: mistralai/Mistral-7B-Instruct-v0.2
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: feedback_p0.1_seed42_level2_raremix
|
| 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 |
+
# feedback_p0.1_seed42_level2_raremix
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.3463
|
| 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: 1e-05
|
| 42 |
+
- train_batch_size: 2
|
| 43 |
+
- eval_batch_size: 1
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 4
|
| 47 |
+
- gradient_accumulation_steps: 4
|
| 48 |
+
- total_train_batch_size: 32
|
| 49 |
+
- total_eval_batch_size: 4
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: cosine
|
| 52 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 53 |
+
- num_epochs: 1
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 58 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 59 |
+
| 0.3509 | 0.9997 | 931 | 0.3463 |
|
| 60 |
+
|
| 61 |
+
|
| 62 |
+
### Framework versions
|
| 63 |
+
|
| 64 |
+
- Transformers 4.43.4
|
| 65 |
+
- Pytorch 2.3.1+cu121
|
| 66 |
+
- Datasets 2.19.1
|
| 67 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9997315436241611,
|
| 3 |
+
"total_flos": 194880114524160.0,
|
| 4 |
+
"train_loss": 0.5085661182086014,
|
| 5 |
+
"train_runtime": 21123.8288,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.41,
|
| 8 |
+
"train_steps_per_second": 0.044
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.43.4"
|
| 6 |
+
}
|
runs/Sep09_09-31-09_COE-CS-sv003/events.out.tfevents.1725874574.COE-CS-sv003.742544.0
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 sha256:0a8d0f976a81ffde932076b669d3faba8d56cd30915e9be0fd57fe4bc8362c6c
|
| 3 |
+
size 45210
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9997315436241611,
|
| 3 |
+
"total_flos": 194880114524160.0,
|
| 4 |
+
"train_loss": 0.5085661182086014,
|
| 5 |
+
"train_runtime": 21123.8288,
|
| 6 |
+
"train_samples": 98952,
|
| 7 |
+
"train_samples_per_second": 1.41,
|
| 8 |
+
"train_steps_per_second": 0.044
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1359 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9997315436241611,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 931,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0010738255033557046,
|
| 13 |
+
"grad_norm": 23.23349641925915,
|
| 14 |
+
"learning_rate": 1.0638297872340426e-07,
|
| 15 |
+
"loss": 1.3224,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005369127516778523,
|
| 20 |
+
"grad_norm": 20.864176480583073,
|
| 21 |
+
"learning_rate": 5.319148936170213e-07,
|
| 22 |
+
"loss": 1.3084,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010738255033557046,
|
| 27 |
+
"grad_norm": 8.533873275523032,
|
| 28 |
+
"learning_rate": 1.0638297872340427e-06,
|
| 29 |
+
"loss": 1.1987,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.016107382550335572,
|
| 34 |
+
"grad_norm": 7.665296900301775,
|
| 35 |
+
"learning_rate": 1.595744680851064e-06,
|
| 36 |
+
"loss": 1.0267,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.021476510067114093,
|
| 41 |
+
"grad_norm": 2.9761601522815546,
|
| 42 |
+
"learning_rate": 2.1276595744680853e-06,
|
| 43 |
+
"loss": 0.9131,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.026845637583892617,
|
| 48 |
+
"grad_norm": 2.433875752280275,
|
| 49 |
+
"learning_rate": 2.6595744680851065e-06,
|
| 50 |
+
"loss": 0.8629,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.032214765100671144,
|
| 55 |
+
"grad_norm": 2.309359589452698,
|
| 56 |
+
"learning_rate": 3.191489361702128e-06,
|
| 57 |
+
"loss": 0.8432,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03758389261744966,
|
| 62 |
+
"grad_norm": 2.3739666896282,
|
| 63 |
+
"learning_rate": 3.723404255319149e-06,
|
| 64 |
+
"loss": 0.8237,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.042953020134228186,
|
| 69 |
+
"grad_norm": 2.411414400789539,
|
| 70 |
+
"learning_rate": 4.255319148936171e-06,
|
| 71 |
+
"loss": 0.8004,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04832214765100671,
|
| 76 |
+
"grad_norm": 2.2779093841935176,
|
| 77 |
+
"learning_rate": 4.787234042553192e-06,
|
| 78 |
+
"loss": 0.7798,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.053691275167785234,
|
| 83 |
+
"grad_norm": 2.390464966558602,
|
| 84 |
+
"learning_rate": 5.319148936170213e-06,
|
| 85 |
+
"loss": 0.777,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.05906040268456376,
|
| 90 |
+
"grad_norm": 2.30705290045996,
|
| 91 |
+
"learning_rate": 5.851063829787235e-06,
|
| 92 |
+
"loss": 0.7534,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.06442953020134229,
|
| 97 |
+
"grad_norm": 2.432088997553532,
|
| 98 |
+
"learning_rate": 6.382978723404256e-06,
|
| 99 |
+
"loss": 0.7363,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.0697986577181208,
|
| 104 |
+
"grad_norm": 2.231418575279909,
|
| 105 |
+
"learning_rate": 6.914893617021278e-06,
|
| 106 |
+
"loss": 0.7258,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.07516778523489932,
|
| 111 |
+
"grad_norm": 2.3360484244386015,
|
| 112 |
+
"learning_rate": 7.446808510638298e-06,
|
| 113 |
+
"loss": 0.7239,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.08053691275167785,
|
| 118 |
+
"grad_norm": 2.2583495209303805,
|
| 119 |
+
"learning_rate": 7.97872340425532e-06,
|
| 120 |
+
"loss": 0.7158,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08590604026845637,
|
| 125 |
+
"grad_norm": 2.166632602977575,
|
| 126 |
+
"learning_rate": 8.510638297872341e-06,
|
| 127 |
+
"loss": 0.698,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0912751677852349,
|
| 132 |
+
"grad_norm": 2.261852982800451,
|
| 133 |
+
"learning_rate": 9.042553191489362e-06,
|
| 134 |
+
"loss": 0.6978,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.09664429530201342,
|
| 139 |
+
"grad_norm": 2.2449184892395593,
|
| 140 |
+
"learning_rate": 9.574468085106385e-06,
|
| 141 |
+
"loss": 0.6914,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.10201342281879194,
|
| 146 |
+
"grad_norm": 2.6796428545722906,
|
| 147 |
+
"learning_rate": 9.999964780082996e-06,
|
| 148 |
+
"loss": 0.6925,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.10738255033557047,
|
| 153 |
+
"grad_norm": 2.445223053748917,
|
| 154 |
+
"learning_rate": 9.998732135085665e-06,
|
| 155 |
+
"loss": 0.6894,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.11275167785234899,
|
| 160 |
+
"grad_norm": 2.101448463042792,
|
| 161 |
+
"learning_rate": 9.995738990383743e-06,
|
| 162 |
+
"loss": 0.6842,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.11812080536912752,
|
| 167 |
+
"grad_norm": 2.375756679579466,
|
| 168 |
+
"learning_rate": 9.990986400130607e-06,
|
| 169 |
+
"loss": 0.6881,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.12348993288590604,
|
| 174 |
+
"grad_norm": 2.0584289730843675,
|
| 175 |
+
"learning_rate": 9.984476038137437e-06,
|
| 176 |
+
"loss": 0.6789,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.12885906040268458,
|
| 181 |
+
"grad_norm": 2.1097466944401946,
|
| 182 |
+
"learning_rate": 9.97621019728372e-06,
|
| 183 |
+
"loss": 0.6794,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.1342281879194631,
|
| 188 |
+
"grad_norm": 2.1000573026325813,
|
| 189 |
+
"learning_rate": 9.966191788709716e-06,
|
| 190 |
+
"loss": 0.6622,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.1395973154362416,
|
| 195 |
+
"grad_norm": 2.21126026156604,
|
| 196 |
+
"learning_rate": 9.954424340791195e-06,
|
| 197 |
+
"loss": 0.6659,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.14496644295302014,
|
| 202 |
+
"grad_norm": 2.233783388990834,
|
| 203 |
+
"learning_rate": 9.940911997896774e-06,
|
| 204 |
+
"loss": 0.6708,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.15033557046979865,
|
| 209 |
+
"grad_norm": 2.3220703167361068,
|
| 210 |
+
"learning_rate": 9.925659518928316e-06,
|
| 211 |
+
"loss": 0.6546,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.15570469798657718,
|
| 216 |
+
"grad_norm": 2.1406068535723626,
|
| 217 |
+
"learning_rate": 9.908672275644898e-06,
|
| 218 |
+
"loss": 0.6606,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1610738255033557,
|
| 223 |
+
"grad_norm": 1.9796338120208212,
|
| 224 |
+
"learning_rate": 9.889956250770933e-06,
|
| 225 |
+
"loss": 0.6543,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.16644295302013423,
|
| 230 |
+
"grad_norm": 2.038004062758435,
|
| 231 |
+
"learning_rate": 9.86951803588912e-06,
|
| 232 |
+
"loss": 0.639,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.17181208053691274,
|
| 237 |
+
"grad_norm": 1.9023679569925318,
|
| 238 |
+
"learning_rate": 9.847364829118963e-06,
|
| 239 |
+
"loss": 0.6394,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.17718120805369128,
|
| 244 |
+
"grad_norm": 1.908984775733616,
|
| 245 |
+
"learning_rate": 9.82350443258166e-06,
|
| 246 |
+
"loss": 0.6449,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1825503355704698,
|
| 251 |
+
"grad_norm": 2.010057009992793,
|
| 252 |
+
"learning_rate": 9.797945249652295e-06,
|
| 253 |
+
"loss": 0.6417,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.18791946308724833,
|
| 258 |
+
"grad_norm": 2.2539350542085006,
|
| 259 |
+
"learning_rate": 9.770696282000245e-06,
|
| 260 |
+
"loss": 0.6333,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.19328859060402684,
|
| 265 |
+
"grad_norm": 1.9065312726735626,
|
| 266 |
+
"learning_rate": 9.741767126418898e-06,
|
| 267 |
+
"loss": 0.6441,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.19865771812080538,
|
| 272 |
+
"grad_norm": 2.0928470931750023,
|
| 273 |
+
"learning_rate": 9.711167971445766e-06,
|
| 274 |
+
"loss": 0.6213,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.2040268456375839,
|
| 279 |
+
"grad_norm": 2.006004562606479,
|
| 280 |
+
"learning_rate": 9.67890959377418e-06,
|
| 281 |
+
"loss": 0.6278,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.20939597315436242,
|
| 286 |
+
"grad_norm": 1.9301686303505856,
|
| 287 |
+
"learning_rate": 9.645003354457872e-06,
|
| 288 |
+
"loss": 0.6325,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.21476510067114093,
|
| 293 |
+
"grad_norm": 2.038658207178946,
|
| 294 |
+
"learning_rate": 9.60946119490972e-06,
|
| 295 |
+
"loss": 0.6361,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.22013422818791947,
|
| 300 |
+
"grad_norm": 2.2159213381686604,
|
| 301 |
+
"learning_rate": 9.57229563269612e-06,
|
| 302 |
+
"loss": 0.619,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.22550335570469798,
|
| 307 |
+
"grad_norm": 2.2179012384166685,
|
| 308 |
+
"learning_rate": 9.533519757128426e-06,
|
| 309 |
+
"loss": 0.6296,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.23087248322147652,
|
| 314 |
+
"grad_norm": 1.9118609526745118,
|
| 315 |
+
"learning_rate": 9.49314722465304e-06,
|
| 316 |
+
"loss": 0.6172,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.23624161073825503,
|
| 321 |
+
"grad_norm": 2.0780299928973753,
|
| 322 |
+
"learning_rate": 9.451192254041759e-06,
|
| 323 |
+
"loss": 0.6238,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.24161073825503357,
|
| 328 |
+
"grad_norm": 1.880745326989776,
|
| 329 |
+
"learning_rate": 9.407669621384073e-06,
|
| 330 |
+
"loss": 0.603,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.24697986577181208,
|
| 335 |
+
"grad_norm": 2.0325599687685325,
|
| 336 |
+
"learning_rate": 9.362594654883185e-06,
|
| 337 |
+
"loss": 0.6098,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.2523489932885906,
|
| 342 |
+
"grad_norm": 2.1114795284546006,
|
| 343 |
+
"learning_rate": 9.31598322945759e-06,
|
| 344 |
+
"loss": 0.5996,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.25771812080536916,
|
| 349 |
+
"grad_norm": 2.147915062806659,
|
| 350 |
+
"learning_rate": 9.267851761150092e-06,
|
| 351 |
+
"loss": 0.5922,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.26308724832214764,
|
| 356 |
+
"grad_norm": 1.8963761596238329,
|
| 357 |
+
"learning_rate": 9.218217201346251e-06,
|
| 358 |
+
"loss": 0.5977,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.2684563758389262,
|
| 363 |
+
"grad_norm": 2.0596529484621406,
|
| 364 |
+
"learning_rate": 9.167097030804289e-06,
|
| 365 |
+
"loss": 0.6019,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2738255033557047,
|
| 370 |
+
"grad_norm": 1.9613750119136377,
|
| 371 |
+
"learning_rate": 9.114509253498554e-06,
|
| 372 |
+
"loss": 0.5823,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2791946308724832,
|
| 377 |
+
"grad_norm": 1.9493095780956693,
|
| 378 |
+
"learning_rate": 9.060472390278717e-06,
|
| 379 |
+
"loss": 0.5895,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.28456375838926173,
|
| 384 |
+
"grad_norm": 1.9649255622722053,
|
| 385 |
+
"learning_rate": 9.005005472346923e-06,
|
| 386 |
+
"loss": 0.5861,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.28993288590604027,
|
| 391 |
+
"grad_norm": 1.916731628798239,
|
| 392 |
+
"learning_rate": 8.948128034555212e-06,
|
| 393 |
+
"loss": 0.5741,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2953020134228188,
|
| 398 |
+
"grad_norm": 1.9882471136788202,
|
| 399 |
+
"learning_rate": 8.889860108525544e-06,
|
| 400 |
+
"loss": 0.5891,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3006711409395973,
|
| 405 |
+
"grad_norm": 1.8657052588450402,
|
| 406 |
+
"learning_rate": 8.83022221559489e-06,
|
| 407 |
+
"loss": 0.5662,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.30604026845637583,
|
| 412 |
+
"grad_norm": 1.8876272358030999,
|
| 413 |
+
"learning_rate": 8.76923535958783e-06,
|
| 414 |
+
"loss": 0.5611,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.31140939597315437,
|
| 419 |
+
"grad_norm": 1.9352681945969277,
|
| 420 |
+
"learning_rate": 8.706921019419237e-06,
|
| 421 |
+
"loss": 0.5703,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.3167785234899329,
|
| 426 |
+
"grad_norm": 2.061655573531995,
|
| 427 |
+
"learning_rate": 8.643301141529619e-06,
|
| 428 |
+
"loss": 0.5704,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.3221476510067114,
|
| 433 |
+
"grad_norm": 1.9955648734094908,
|
| 434 |
+
"learning_rate": 8.578398132155846e-06,
|
| 435 |
+
"loss": 0.5646,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3275167785234899,
|
| 440 |
+
"grad_norm": 1.9172018423696509,
|
| 441 |
+
"learning_rate": 8.512234849439887e-06,
|
| 442 |
+
"loss": 0.5603,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.33288590604026846,
|
| 447 |
+
"grad_norm": 2.748943489124992,
|
| 448 |
+
"learning_rate": 8.444834595378434e-06,
|
| 449 |
+
"loss": 0.5635,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.338255033557047,
|
| 454 |
+
"grad_norm": 2.031805396620389,
|
| 455 |
+
"learning_rate": 8.376221107616187e-06,
|
| 456 |
+
"loss": 0.56,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.3436241610738255,
|
| 461 |
+
"grad_norm": 2.310478182840465,
|
| 462 |
+
"learning_rate": 8.306418551085707e-06,
|
| 463 |
+
"loss": 0.5518,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.348993288590604,
|
| 468 |
+
"grad_norm": 2.143453110103673,
|
| 469 |
+
"learning_rate": 8.23545150949679e-06,
|
| 470 |
+
"loss": 0.545,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.35436241610738256,
|
| 475 |
+
"grad_norm": 2.074828415385107,
|
| 476 |
+
"learning_rate": 8.163344976678342e-06,
|
| 477 |
+
"loss": 0.5417,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3597315436241611,
|
| 482 |
+
"grad_norm": 1.95080697885721,
|
| 483 |
+
"learning_rate": 8.090124347775837e-06,
|
| 484 |
+
"loss": 0.5461,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.3651006711409396,
|
| 489 |
+
"grad_norm": 1.8940424474498196,
|
| 490 |
+
"learning_rate": 8.0158154103074e-06,
|
| 491 |
+
"loss": 0.5415,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.3704697986577181,
|
| 496 |
+
"grad_norm": 1.9899368306488876,
|
| 497 |
+
"learning_rate": 7.940444335081733e-06,
|
| 498 |
+
"loss": 0.5334,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.37583892617449666,
|
| 503 |
+
"grad_norm": 3.3190577946485074,
|
| 504 |
+
"learning_rate": 7.864037666981037e-06,
|
| 505 |
+
"loss": 0.5274,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3812080536912752,
|
| 510 |
+
"grad_norm": 1.9132824928499255,
|
| 511 |
+
"learning_rate": 7.786622315612182e-06,
|
| 512 |
+
"loss": 0.5395,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3865771812080537,
|
| 517 |
+
"grad_norm": 1.9703232913555617,
|
| 518 |
+
"learning_rate": 7.708225545829446e-06,
|
| 519 |
+
"loss": 0.5213,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3919463087248322,
|
| 524 |
+
"grad_norm": 2.076133929106472,
|
| 525 |
+
"learning_rate": 7.6288749681321115e-06,
|
| 526 |
+
"loss": 0.5356,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.39731543624161075,
|
| 531 |
+
"grad_norm": 2.0858297493245637,
|
| 532 |
+
"learning_rate": 7.548598528940354e-06,
|
| 533 |
+
"loss": 0.5286,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.40268456375838924,
|
| 538 |
+
"grad_norm": 1.9498254037168414,
|
| 539 |
+
"learning_rate": 7.4674245007528135e-06,
|
| 540 |
+
"loss": 0.5217,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.4080536912751678,
|
| 545 |
+
"grad_norm": 1.9718043172163682,
|
| 546 |
+
"learning_rate": 7.385381472189321e-06,
|
| 547 |
+
"loss": 0.5185,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.4134228187919463,
|
| 552 |
+
"grad_norm": 2.0714251046483136,
|
| 553 |
+
"learning_rate": 7.302498337922293e-06,
|
| 554 |
+
"loss": 0.5212,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.41879194630872485,
|
| 559 |
+
"grad_norm": 2.059741936941689,
|
| 560 |
+
"learning_rate": 7.218804288500343e-06,
|
| 561 |
+
"loss": 0.5068,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.42416107382550333,
|
| 566 |
+
"grad_norm": 2.0677503384082656,
|
| 567 |
+
"learning_rate": 7.134328800067684e-06,
|
| 568 |
+
"loss": 0.5128,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.42953020134228187,
|
| 573 |
+
"grad_norm": 2.1270598611660008,
|
| 574 |
+
"learning_rate": 7.049101623982938e-06,
|
| 575 |
+
"loss": 0.5056,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.4348993288590604,
|
| 580 |
+
"grad_norm": 1.7730274826372387,
|
| 581 |
+
"learning_rate": 6.963152776341044e-06,
|
| 582 |
+
"loss": 0.5007,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.44026845637583895,
|
| 587 |
+
"grad_norm": 2.092171850544149,
|
| 588 |
+
"learning_rate": 6.876512527401897e-06,
|
| 589 |
+
"loss": 0.4969,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.44563758389261743,
|
| 594 |
+
"grad_norm": 1.8511960894357482,
|
| 595 |
+
"learning_rate": 6.789211390929497e-06,
|
| 596 |
+
"loss": 0.5069,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.45100671140939597,
|
| 601 |
+
"grad_norm": 2.003279508110089,
|
| 602 |
+
"learning_rate": 6.701280113445324e-06,
|
| 603 |
+
"loss": 0.4868,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.4563758389261745,
|
| 608 |
+
"grad_norm": 1.9910852122088196,
|
| 609 |
+
"learning_rate": 6.6127496633997475e-06,
|
| 610 |
+
"loss": 0.4856,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.46174496644295304,
|
| 615 |
+
"grad_norm": 2.0142055270514425,
|
| 616 |
+
"learning_rate": 6.523651220265269e-06,
|
| 617 |
+
"loss": 0.4942,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.4671140939597315,
|
| 622 |
+
"grad_norm": 2.102485865135929,
|
| 623 |
+
"learning_rate": 6.434016163555452e-06,
|
| 624 |
+
"loss": 0.4848,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.47248322147651006,
|
| 629 |
+
"grad_norm": 2.1181467567869623,
|
| 630 |
+
"learning_rate": 6.343876061773385e-06,
|
| 631 |
+
"loss": 0.4886,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.4778523489932886,
|
| 636 |
+
"grad_norm": 2.294633696659754,
|
| 637 |
+
"learning_rate": 6.2532626612936035e-06,
|
| 638 |
+
"loss": 0.4862,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.48322147651006714,
|
| 643 |
+
"grad_norm": 1.9401139366264295,
|
| 644 |
+
"learning_rate": 6.162207875181354e-06,
|
| 645 |
+
"loss": 0.4834,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.4885906040268456,
|
| 650 |
+
"grad_norm": 1.8938897786171476,
|
| 651 |
+
"learning_rate": 6.070743771953157e-06,
|
| 652 |
+
"loss": 0.4769,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.49395973154362416,
|
| 657 |
+
"grad_norm": 2.035803586489251,
|
| 658 |
+
"learning_rate": 5.978902564282616e-06,
|
| 659 |
+
"loss": 0.4819,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.4993288590604027,
|
| 664 |
+
"grad_norm": 1.8943417586526035,
|
| 665 |
+
"learning_rate": 5.886716597655472e-06,
|
| 666 |
+
"loss": 0.4759,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.5046979865771812,
|
| 671 |
+
"grad_norm": 1.8388956694087875,
|
| 672 |
+
"learning_rate": 5.7942183389778536e-06,
|
| 673 |
+
"loss": 0.4683,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.5100671140939598,
|
| 678 |
+
"grad_norm": 1.9348760408299013,
|
| 679 |
+
"learning_rate": 5.701440365141799e-06,
|
| 680 |
+
"loss": 0.4644,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5154362416107383,
|
| 685 |
+
"grad_norm": 2.0489951854475796,
|
| 686 |
+
"learning_rate": 5.608415351552014e-06,
|
| 687 |
+
"loss": 0.4662,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.5208053691275167,
|
| 692 |
+
"grad_norm": 1.9191064778476148,
|
| 693 |
+
"learning_rate": 5.515176060617945e-06,
|
| 694 |
+
"loss": 0.4629,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5261744966442953,
|
| 699 |
+
"grad_norm": 2.1725377637259924,
|
| 700 |
+
"learning_rate": 5.421755330215223e-06,
|
| 701 |
+
"loss": 0.4562,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.5315436241610738,
|
| 706 |
+
"grad_norm": 1.9581424804999248,
|
| 707 |
+
"learning_rate": 5.328186062120509e-06,
|
| 708 |
+
"loss": 0.4635,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5369127516778524,
|
| 713 |
+
"grad_norm": 1.8929231268679836,
|
| 714 |
+
"learning_rate": 5.23450121042383e-06,
|
| 715 |
+
"loss": 0.454,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.5422818791946309,
|
| 720 |
+
"grad_norm": 1.9802801188021257,
|
| 721 |
+
"learning_rate": 5.140733769922525e-06,
|
| 722 |
+
"loss": 0.4641,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.5476510067114094,
|
| 727 |
+
"grad_norm": 2.008695527313478,
|
| 728 |
+
"learning_rate": 5.0469167645008245e-06,
|
| 729 |
+
"loss": 0.4482,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.553020134228188,
|
| 734 |
+
"grad_norm": 2.010797136402869,
|
| 735 |
+
"learning_rate": 4.953083235499177e-06,
|
| 736 |
+
"loss": 0.4407,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5583892617449664,
|
| 741 |
+
"grad_norm": 1.9094101117542013,
|
| 742 |
+
"learning_rate": 4.859266230077474e-06,
|
| 743 |
+
"loss": 0.4479,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5637583892617449,
|
| 748 |
+
"grad_norm": 1.9310237131735821,
|
| 749 |
+
"learning_rate": 4.7654987895761705e-06,
|
| 750 |
+
"loss": 0.4421,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5691275167785235,
|
| 755 |
+
"grad_norm": 1.9040107482286928,
|
| 756 |
+
"learning_rate": 4.671813937879494e-06,
|
| 757 |
+
"loss": 0.446,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.574496644295302,
|
| 762 |
+
"grad_norm": 2.0403382751342387,
|
| 763 |
+
"learning_rate": 4.5782446697847775e-06,
|
| 764 |
+
"loss": 0.4469,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5798657718120805,
|
| 769 |
+
"grad_norm": 1.827855233692127,
|
| 770 |
+
"learning_rate": 4.484823939382056e-06,
|
| 771 |
+
"loss": 0.4349,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5852348993288591,
|
| 776 |
+
"grad_norm": 1.8653574780339666,
|
| 777 |
+
"learning_rate": 4.391584648447989e-06,
|
| 778 |
+
"loss": 0.4288,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5906040268456376,
|
| 783 |
+
"grad_norm": 1.986995981845508,
|
| 784 |
+
"learning_rate": 4.298559634858202e-06,
|
| 785 |
+
"loss": 0.4372,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5959731543624162,
|
| 790 |
+
"grad_norm": 1.9833632848096547,
|
| 791 |
+
"learning_rate": 4.205781661022146e-06,
|
| 792 |
+
"loss": 0.4332,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.6013422818791946,
|
| 797 |
+
"grad_norm": 1.9751140496991637,
|
| 798 |
+
"learning_rate": 4.1132834023445304e-06,
|
| 799 |
+
"loss": 0.4478,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.6067114093959731,
|
| 804 |
+
"grad_norm": 2.0032559169326456,
|
| 805 |
+
"learning_rate": 4.021097435717386e-06,
|
| 806 |
+
"loss": 0.4221,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.6120805369127517,
|
| 811 |
+
"grad_norm": 1.9079295664632885,
|
| 812 |
+
"learning_rate": 3.929256228046845e-06,
|
| 813 |
+
"loss": 0.4157,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.6174496644295302,
|
| 818 |
+
"grad_norm": 1.8781628903110001,
|
| 819 |
+
"learning_rate": 3.837792124818647e-06,
|
| 820 |
+
"loss": 0.4277,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.6228187919463087,
|
| 825 |
+
"grad_norm": 1.7915826382801951,
|
| 826 |
+
"learning_rate": 3.7467373387063973e-06,
|
| 827 |
+
"loss": 0.4241,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.6281879194630873,
|
| 832 |
+
"grad_norm": 1.8613980836008768,
|
| 833 |
+
"learning_rate": 3.656123938226618e-06,
|
| 834 |
+
"loss": 0.4153,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.6335570469798658,
|
| 839 |
+
"grad_norm": 1.8036271371193608,
|
| 840 |
+
"learning_rate": 3.5659838364445505e-06,
|
| 841 |
+
"loss": 0.4087,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6389261744966444,
|
| 846 |
+
"grad_norm": 1.876894262488646,
|
| 847 |
+
"learning_rate": 3.476348779734732e-06,
|
| 848 |
+
"loss": 0.4171,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6442953020134228,
|
| 853 |
+
"grad_norm": 2.005365914493537,
|
| 854 |
+
"learning_rate": 3.387250336600254e-06,
|
| 855 |
+
"loss": 0.4117,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6496644295302013,
|
| 860 |
+
"grad_norm": 1.847827057920518,
|
| 861 |
+
"learning_rate": 3.298719886554677e-06,
|
| 862 |
+
"loss": 0.4129,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.6550335570469799,
|
| 867 |
+
"grad_norm": 1.8912137540309444,
|
| 868 |
+
"learning_rate": 3.2107886090705035e-06,
|
| 869 |
+
"loss": 0.4006,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.6604026845637584,
|
| 874 |
+
"grad_norm": 1.8594474550600166,
|
| 875 |
+
"learning_rate": 3.1234874725981045e-06,
|
| 876 |
+
"loss": 0.4043,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.6657718120805369,
|
| 881 |
+
"grad_norm": 1.8497041114366601,
|
| 882 |
+
"learning_rate": 3.036847223658958e-06,
|
| 883 |
+
"loss": 0.4033,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.6711409395973155,
|
| 888 |
+
"grad_norm": 1.960254189224025,
|
| 889 |
+
"learning_rate": 2.950898376017064e-06,
|
| 890 |
+
"loss": 0.4036,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.676510067114094,
|
| 895 |
+
"grad_norm": 2.10629670479599,
|
| 896 |
+
"learning_rate": 2.865671199932318e-06,
|
| 897 |
+
"loss": 0.4044,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.6818791946308724,
|
| 902 |
+
"grad_norm": 1.8824907032670186,
|
| 903 |
+
"learning_rate": 2.781195711499658e-06,
|
| 904 |
+
"loss": 0.4095,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.687248322147651,
|
| 909 |
+
"grad_norm": 1.9655088963405234,
|
| 910 |
+
"learning_rate": 2.697501662077707e-06,
|
| 911 |
+
"loss": 0.4021,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.6926174496644295,
|
| 916 |
+
"grad_norm": 1.889034517852432,
|
| 917 |
+
"learning_rate": 2.6146185278106807e-06,
|
| 918 |
+
"loss": 0.3825,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.697986577181208,
|
| 923 |
+
"grad_norm": 1.781359751443117,
|
| 924 |
+
"learning_rate": 2.5325754992471886e-06,
|
| 925 |
+
"loss": 0.403,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.7033557046979866,
|
| 930 |
+
"grad_norm": 2.124217994937085,
|
| 931 |
+
"learning_rate": 2.4514014710596467e-06,
|
| 932 |
+
"loss": 0.3938,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.7087248322147651,
|
| 937 |
+
"grad_norm": 2.0049604807831636,
|
| 938 |
+
"learning_rate": 2.3711250318678906e-06,
|
| 939 |
+
"loss": 0.3943,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.7140939597315437,
|
| 944 |
+
"grad_norm": 1.8963570644553214,
|
| 945 |
+
"learning_rate": 2.2917744541705544e-06,
|
| 946 |
+
"loss": 0.3886,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.7194630872483222,
|
| 951 |
+
"grad_norm": 1.9231403525555266,
|
| 952 |
+
"learning_rate": 2.2133776843878185e-06,
|
| 953 |
+
"loss": 0.3918,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7248322147651006,
|
| 958 |
+
"grad_norm": 1.9140311429947394,
|
| 959 |
+
"learning_rate": 2.1359623330189655e-06,
|
| 960 |
+
"loss": 0.3924,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7302013422818792,
|
| 965 |
+
"grad_norm": 1.790471011384921,
|
| 966 |
+
"learning_rate": 2.059555664918268e-06,
|
| 967 |
+
"loss": 0.3861,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.7355704697986577,
|
| 972 |
+
"grad_norm": 1.8245278498502604,
|
| 973 |
+
"learning_rate": 1.9841845896926022e-06,
|
| 974 |
+
"loss": 0.3837,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.7409395973154362,
|
| 979 |
+
"grad_norm": 1.835597641802119,
|
| 980 |
+
"learning_rate": 1.9098756522241634e-06,
|
| 981 |
+
"loss": 0.3833,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.7463087248322148,
|
| 986 |
+
"grad_norm": 1.7733798731081618,
|
| 987 |
+
"learning_rate": 1.8366550233216584e-06,
|
| 988 |
+
"loss": 0.3859,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.7516778523489933,
|
| 993 |
+
"grad_norm": 1.8909772302753176,
|
| 994 |
+
"learning_rate": 1.7645484905032129e-06,
|
| 995 |
+
"loss": 0.3725,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.7570469798657719,
|
| 1000 |
+
"grad_norm": 1.8395377457755753,
|
| 1001 |
+
"learning_rate": 1.6935814489142937e-06,
|
| 1002 |
+
"loss": 0.375,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.7624161073825504,
|
| 1007 |
+
"grad_norm": 2.004189882277981,
|
| 1008 |
+
"learning_rate": 1.6237788923838149e-06,
|
| 1009 |
+
"loss": 0.3745,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.7677852348993288,
|
| 1014 |
+
"grad_norm": 1.8493982881721263,
|
| 1015 |
+
"learning_rate": 1.555165404621567e-06,
|
| 1016 |
+
"loss": 0.3851,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.7731543624161074,
|
| 1021 |
+
"grad_norm": 1.8731392963530693,
|
| 1022 |
+
"learning_rate": 1.487765150560116e-06,
|
| 1023 |
+
"loss": 0.3719,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.7785234899328859,
|
| 1028 |
+
"grad_norm": 1.800997454509494,
|
| 1029 |
+
"learning_rate": 1.4216018678441558e-06,
|
| 1030 |
+
"loss": 0.3735,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.7838926174496644,
|
| 1035 |
+
"grad_norm": 1.858224940166772,
|
| 1036 |
+
"learning_rate": 1.3566988584703817e-06,
|
| 1037 |
+
"loss": 0.3674,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.789261744966443,
|
| 1042 |
+
"grad_norm": 1.8306556119241966,
|
| 1043 |
+
"learning_rate": 1.293078980580766e-06,
|
| 1044 |
+
"loss": 0.3731,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.7946308724832215,
|
| 1049 |
+
"grad_norm": 1.7780036225720621,
|
| 1050 |
+
"learning_rate": 1.2307646404121692e-06,
|
| 1051 |
+
"loss": 0.362,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.8,
|
| 1056 |
+
"grad_norm": 1.9057965494467726,
|
| 1057 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 1058 |
+
"loss": 0.3711,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.8053691275167785,
|
| 1063 |
+
"grad_norm": 1.7752487152814125,
|
| 1064 |
+
"learning_rate": 1.1101398914744565e-06,
|
| 1065 |
+
"loss": 0.3644,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.810738255033557,
|
| 1070 |
+
"grad_norm": 1.8233485934046587,
|
| 1071 |
+
"learning_rate": 1.0518719654447896e-06,
|
| 1072 |
+
"loss": 0.3635,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.8161073825503355,
|
| 1077 |
+
"grad_norm": 1.7956083349387602,
|
| 1078 |
+
"learning_rate": 9.949945276530782e-07,
|
| 1079 |
+
"loss": 0.361,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.8214765100671141,
|
| 1084 |
+
"grad_norm": 1.7050898287128047,
|
| 1085 |
+
"learning_rate": 9.395276097212841e-07,
|
| 1086 |
+
"loss": 0.3632,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.8268456375838926,
|
| 1091 |
+
"grad_norm": 1.8682808768340036,
|
| 1092 |
+
"learning_rate": 8.854907465014479e-07,
|
| 1093 |
+
"loss": 0.3679,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.8322147651006712,
|
| 1098 |
+
"grad_norm": 1.7567008107111888,
|
| 1099 |
+
"learning_rate": 8.329029691957124e-07,
|
| 1100 |
+
"loss": 0.3663,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.8375838926174497,
|
| 1105 |
+
"grad_norm": 1.8090009300620273,
|
| 1106 |
+
"learning_rate": 7.817827986537508e-07,
|
| 1107 |
+
"loss": 0.3552,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.8429530201342282,
|
| 1112 |
+
"grad_norm": 1.8320510025717955,
|
| 1113 |
+
"learning_rate": 7.321482388499096e-07,
|
| 1114 |
+
"loss": 0.3595,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.8483221476510067,
|
| 1119 |
+
"grad_norm": 1.80583578265842,
|
| 1120 |
+
"learning_rate": 6.840167705424106e-07,
|
| 1121 |
+
"loss": 0.3652,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8536912751677852,
|
| 1126 |
+
"grad_norm": 1.8327102347329405,
|
| 1127 |
+
"learning_rate": 6.374053451168166e-07,
|
| 1128 |
+
"loss": 0.3561,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.8590604026845637,
|
| 1133 |
+
"grad_norm": 1.8311020050096911,
|
| 1134 |
+
"learning_rate": 5.92330378615929e-07,
|
| 1135 |
+
"loss": 0.3585,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.8644295302013423,
|
| 1140 |
+
"grad_norm": 1.8055339358917153,
|
| 1141 |
+
"learning_rate": 5.488077459582425e-07,
|
| 1142 |
+
"loss": 0.3593,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.8697986577181208,
|
| 1147 |
+
"grad_norm": 1.8142078683309888,
|
| 1148 |
+
"learning_rate": 5.068527753469604e-07,
|
| 1149 |
+
"loss": 0.3556,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.8751677852348994,
|
| 1154 |
+
"grad_norm": 1.8006022957772634,
|
| 1155 |
+
"learning_rate": 4.664802428715753e-07,
|
| 1156 |
+
"loss": 0.3501,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.8805369127516779,
|
| 1161 |
+
"grad_norm": 1.8115511806620141,
|
| 1162 |
+
"learning_rate": 4.2770436730388166e-07,
|
| 1163 |
+
"loss": 0.3555,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.8859060402684564,
|
| 1168 |
+
"grad_norm": 1.8145406688101844,
|
| 1169 |
+
"learning_rate": 3.9053880509028086e-07,
|
| 1170 |
+
"loss": 0.345,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.8912751677852349,
|
| 1175 |
+
"grad_norm": 1.7552883220570528,
|
| 1176 |
+
"learning_rate": 3.549966455421305e-07,
|
| 1177 |
+
"loss": 0.3484,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.8966442953020134,
|
| 1182 |
+
"grad_norm": 1.7075911431534658,
|
| 1183 |
+
"learning_rate": 3.2109040622582186e-07,
|
| 1184 |
+
"loss": 0.3588,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.9020134228187919,
|
| 1189 |
+
"grad_norm": 1.6934879805875171,
|
| 1190 |
+
"learning_rate": 2.8883202855423676e-07,
|
| 1191 |
+
"loss": 0.3583,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.9073825503355705,
|
| 1196 |
+
"grad_norm": 1.77775983103448,
|
| 1197 |
+
"learning_rate": 2.582328735811029e-07,
|
| 1198 |
+
"loss": 0.351,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.912751677852349,
|
| 1203 |
+
"grad_norm": 1.8618113993543914,
|
| 1204 |
+
"learning_rate": 2.2930371799975593e-07,
|
| 1205 |
+
"loss": 0.3491,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.9181208053691275,
|
| 1210 |
+
"grad_norm": 1.858661836026022,
|
| 1211 |
+
"learning_rate": 2.0205475034770606e-07,
|
| 1212 |
+
"loss": 0.3458,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.9234899328859061,
|
| 1217 |
+
"grad_norm": 1.7009249535049955,
|
| 1218 |
+
"learning_rate": 1.7649556741833995e-07,
|
| 1219 |
+
"loss": 0.3456,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.9288590604026845,
|
| 1224 |
+
"grad_norm": 1.792030192988469,
|
| 1225 |
+
"learning_rate": 1.5263517088103862e-07,
|
| 1226 |
+
"loss": 0.3575,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.934228187919463,
|
| 1231 |
+
"grad_norm": 1.9256857866336536,
|
| 1232 |
+
"learning_rate": 1.304819641108801e-07,
|
| 1233 |
+
"loss": 0.3571,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.9395973154362416,
|
| 1238 |
+
"grad_norm": 1.84021155443979,
|
| 1239 |
+
"learning_rate": 1.1004374922906846e-07,
|
| 1240 |
+
"loss": 0.3478,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.9449664429530201,
|
| 1245 |
+
"grad_norm": 1.706774577248585,
|
| 1246 |
+
"learning_rate": 9.132772435510362e-08,
|
| 1247 |
+
"loss": 0.3459,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.9503355704697987,
|
| 1252 |
+
"grad_norm": 1.737896821789404,
|
| 1253 |
+
"learning_rate": 7.434048107168523e-08,
|
| 1254 |
+
"loss": 0.3487,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.9557046979865772,
|
| 1259 |
+
"grad_norm": 1.754018893680602,
|
| 1260 |
+
"learning_rate": 5.908800210322696e-08,
|
| 1261 |
+
"loss": 0.3533,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.9610738255033557,
|
| 1266 |
+
"grad_norm": 1.6854396215038576,
|
| 1267 |
+
"learning_rate": 4.55756592088058e-08,
|
| 1268 |
+
"loss": 0.3493,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.9664429530201343,
|
| 1273 |
+
"grad_norm": 1.7570339706613218,
|
| 1274 |
+
"learning_rate": 3.3808211290284886e-08,
|
| 1275 |
+
"loss": 0.3389,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.9718120805369127,
|
| 1280 |
+
"grad_norm": 1.83552844908017,
|
| 1281 |
+
"learning_rate": 2.378980271628195e-08,
|
| 1282 |
+
"loss": 0.3455,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.9771812080536912,
|
| 1287 |
+
"grad_norm": 1.6940562777678623,
|
| 1288 |
+
"learning_rate": 1.552396186256411e-08,
|
| 1289 |
+
"loss": 0.3478,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.9825503355704698,
|
| 1294 |
+
"grad_norm": 1.7576783825666717,
|
| 1295 |
+
"learning_rate": 9.013599869394096e-09,
|
| 1296 |
+
"loss": 0.3393,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.9879194630872483,
|
| 1301 |
+
"grad_norm": 1.7480069996022527,
|
| 1302 |
+
"learning_rate": 4.261009616257638e-09,
|
| 1303 |
+
"loss": 0.3499,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.9932885906040269,
|
| 1308 |
+
"grad_norm": 1.6906969685299382,
|
| 1309 |
+
"learning_rate": 1.2678649143349485e-09,
|
| 1310 |
+
"loss": 0.3528,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.9986577181208054,
|
| 1315 |
+
"grad_norm": 1.77167724060171,
|
| 1316 |
+
"learning_rate": 3.5219917003948003e-11,
|
| 1317 |
+
"loss": 0.3509,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.9997315436241611,
|
| 1322 |
+
"eval_loss": 0.3463418781757355,
|
| 1323 |
+
"eval_runtime": 1.1928,
|
| 1324 |
+
"eval_samples_per_second": 1.677,
|
| 1325 |
+
"eval_steps_per_second": 0.838,
|
| 1326 |
+
"step": 931
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 0.9997315436241611,
|
| 1330 |
+
"step": 931,
|
| 1331 |
+
"total_flos": 194880114524160.0,
|
| 1332 |
+
"train_loss": 0.5085661182086014,
|
| 1333 |
+
"train_runtime": 21123.8288,
|
| 1334 |
+
"train_samples_per_second": 1.41,
|
| 1335 |
+
"train_steps_per_second": 0.044
|
| 1336 |
+
}
|
| 1337 |
+
],
|
| 1338 |
+
"logging_steps": 5,
|
| 1339 |
+
"max_steps": 931,
|
| 1340 |
+
"num_input_tokens_seen": 0,
|
| 1341 |
+
"num_train_epochs": 1,
|
| 1342 |
+
"save_steps": 100,
|
| 1343 |
+
"stateful_callbacks": {
|
| 1344 |
+
"TrainerControl": {
|
| 1345 |
+
"args": {
|
| 1346 |
+
"should_epoch_stop": false,
|
| 1347 |
+
"should_evaluate": false,
|
| 1348 |
+
"should_log": false,
|
| 1349 |
+
"should_save": true,
|
| 1350 |
+
"should_training_stop": true
|
| 1351 |
+
},
|
| 1352 |
+
"attributes": {}
|
| 1353 |
+
}
|
| 1354 |
+
},
|
| 1355 |
+
"total_flos": 194880114524160.0,
|
| 1356 |
+
"train_batch_size": 2,
|
| 1357 |
+
"trial_name": null,
|
| 1358 |
+
"trial_params": null
|
| 1359 |
+
}
|