Model save
Browse files- README.md +67 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- runs/Aug26_09-43-59_COE-CS-sv003/events.out.tfevents.1724665739.COE-CS-sv003.3081243.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1338 -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: direct_0.1p_seed42_level2_static
|
| 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 |
+
# direct_0.1p_seed42_level2_static
|
| 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: 0.2971
|
| 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 |
+
| 0.3383 | 0.9997 | 915 | 0.2971 |
|
| 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": 0.9997268505872713,
|
| 3 |
+
"total_flos": 191530040033280.0,
|
| 4 |
+
"train_loss": 0.4996221627042593,
|
| 5 |
+
"train_runtime": 20787.7504,
|
| 6 |
+
"train_samples": 98857,
|
| 7 |
+
"train_samples_per_second": 1.409,
|
| 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/Aug26_09-43-59_COE-CS-sv003/events.out.tfevents.1724665739.COE-CS-sv003.3081243.0
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:d3223cfc30a5dfd45301caeb35c7514593a7a76aed598d8aa9ffe5f0606ca81a
|
| 3 |
+
size 44565
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9997268505872713,
|
| 3 |
+
"total_flos": 191530040033280.0,
|
| 4 |
+
"train_loss": 0.4996221627042593,
|
| 5 |
+
"train_runtime": 20787.7504,
|
| 6 |
+
"train_samples": 98857,
|
| 7 |
+
"train_samples_per_second": 1.409,
|
| 8 |
+
"train_steps_per_second": 0.044
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1338 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9997268505872713,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 915,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0010925976509150504,
|
| 13 |
+
"grad_norm": 22.730928776382456,
|
| 14 |
+
"learning_rate": 1.0869565217391305e-07,
|
| 15 |
+
"loss": 1.3223,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005462988254575253,
|
| 20 |
+
"grad_norm": 20.891256691927374,
|
| 21 |
+
"learning_rate": 5.434782608695653e-07,
|
| 22 |
+
"loss": 1.3066,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010925976509150505,
|
| 27 |
+
"grad_norm": 8.917157931216027,
|
| 28 |
+
"learning_rate": 1.0869565217391306e-06,
|
| 29 |
+
"loss": 1.1943,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01638896476372576,
|
| 34 |
+
"grad_norm": 7.290952602788098,
|
| 35 |
+
"learning_rate": 1.6304347826086957e-06,
|
| 36 |
+
"loss": 1.0279,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.02185195301830101,
|
| 41 |
+
"grad_norm": 3.006256144496407,
|
| 42 |
+
"learning_rate": 2.173913043478261e-06,
|
| 43 |
+
"loss": 0.9049,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.027314941272876262,
|
| 48 |
+
"grad_norm": 2.371134339375521,
|
| 49 |
+
"learning_rate": 2.7173913043478263e-06,
|
| 50 |
+
"loss": 0.8647,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.03277792952745152,
|
| 55 |
+
"grad_norm": 2.1504800702490865,
|
| 56 |
+
"learning_rate": 3.2608695652173914e-06,
|
| 57 |
+
"loss": 0.833,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03824091778202677,
|
| 62 |
+
"grad_norm": 2.2317280108941,
|
| 63 |
+
"learning_rate": 3.804347826086957e-06,
|
| 64 |
+
"loss": 0.807,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.04370390603660202,
|
| 69 |
+
"grad_norm": 2.2488850758737255,
|
| 70 |
+
"learning_rate": 4.347826086956522e-06,
|
| 71 |
+
"loss": 0.7919,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04916689429117727,
|
| 76 |
+
"grad_norm": 2.3580705798318484,
|
| 77 |
+
"learning_rate": 4.891304347826087e-06,
|
| 78 |
+
"loss": 0.7761,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.054629882545752524,
|
| 83 |
+
"grad_norm": 2.4373495899440414,
|
| 84 |
+
"learning_rate": 5.4347826086956525e-06,
|
| 85 |
+
"loss": 0.7614,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.060092870800327776,
|
| 90 |
+
"grad_norm": 2.4135433324014763,
|
| 91 |
+
"learning_rate": 5.978260869565218e-06,
|
| 92 |
+
"loss": 0.7458,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.06555585905490303,
|
| 97 |
+
"grad_norm": 2.81358704818424,
|
| 98 |
+
"learning_rate": 6.521739130434783e-06,
|
| 99 |
+
"loss": 0.7291,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.07101884730947829,
|
| 104 |
+
"grad_norm": 2.438627231166693,
|
| 105 |
+
"learning_rate": 7.065217391304349e-06,
|
| 106 |
+
"loss": 0.72,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.07648183556405354,
|
| 111 |
+
"grad_norm": 2.1869040708927177,
|
| 112 |
+
"learning_rate": 7.608695652173914e-06,
|
| 113 |
+
"loss": 0.7141,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.08194482381862879,
|
| 118 |
+
"grad_norm": 2.184483160699592,
|
| 119 |
+
"learning_rate": 8.15217391304348e-06,
|
| 120 |
+
"loss": 0.6994,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08740781207320404,
|
| 125 |
+
"grad_norm": 2.3203787824757507,
|
| 126 |
+
"learning_rate": 8.695652173913044e-06,
|
| 127 |
+
"loss": 0.7019,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0928708003277793,
|
| 132 |
+
"grad_norm": 2.3068506646214915,
|
| 133 |
+
"learning_rate": 9.23913043478261e-06,
|
| 134 |
+
"loss": 0.6958,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.09833378858235454,
|
| 139 |
+
"grad_norm": 2.257062695334498,
|
| 140 |
+
"learning_rate": 9.782608695652175e-06,
|
| 141 |
+
"loss": 0.6842,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1037967768369298,
|
| 146 |
+
"grad_norm": 2.400983485932064,
|
| 147 |
+
"learning_rate": 9.999672147992806e-06,
|
| 148 |
+
"loss": 0.6914,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.10925976509150505,
|
| 153 |
+
"grad_norm": 2.324571446077097,
|
| 154 |
+
"learning_rate": 9.997668763647963e-06,
|
| 155 |
+
"loss": 0.6867,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.1147227533460803,
|
| 160 |
+
"grad_norm": 2.137443230070172,
|
| 161 |
+
"learning_rate": 9.993844863856281e-06,
|
| 162 |
+
"loss": 0.689,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.12018574160065555,
|
| 167 |
+
"grad_norm": 2.0564600115869585,
|
| 168 |
+
"learning_rate": 9.988201841560945e-06,
|
| 169 |
+
"loss": 0.6688,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.12564872985523082,
|
| 174 |
+
"grad_norm": 2.139005852684594,
|
| 175 |
+
"learning_rate": 9.980741752362222e-06,
|
| 176 |
+
"loss": 0.6711,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.13111171810980607,
|
| 181 |
+
"grad_norm": 2.530899194975221,
|
| 182 |
+
"learning_rate": 9.971467313768668e-06,
|
| 183 |
+
"loss": 0.6681,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.13657470636438132,
|
| 188 |
+
"grad_norm": 2.1541912419217466,
|
| 189 |
+
"learning_rate": 9.96038190420721e-06,
|
| 190 |
+
"loss": 0.6624,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.14203769461895657,
|
| 195 |
+
"grad_norm": 2.2266155964762646,
|
| 196 |
+
"learning_rate": 9.947489561792475e-06,
|
| 197 |
+
"loss": 0.6678,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.14750068287353182,
|
| 202 |
+
"grad_norm": 2.0151113143541215,
|
| 203 |
+
"learning_rate": 9.932794982855818e-06,
|
| 204 |
+
"loss": 0.6576,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.15296367112810708,
|
| 209 |
+
"grad_norm": 2.122951721690571,
|
| 210 |
+
"learning_rate": 9.916303520234573e-06,
|
| 211 |
+
"loss": 0.6518,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.15842665938268233,
|
| 216 |
+
"grad_norm": 2.073005839186561,
|
| 217 |
+
"learning_rate": 9.898021181322158e-06,
|
| 218 |
+
"loss": 0.6467,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.16388964763725758,
|
| 223 |
+
"grad_norm": 2.122513038916697,
|
| 224 |
+
"learning_rate": 9.877954625879746e-06,
|
| 225 |
+
"loss": 0.6422,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.16935263589183283,
|
| 230 |
+
"grad_norm": 1.938211750086781,
|
| 231 |
+
"learning_rate": 9.8561111636103e-06,
|
| 232 |
+
"loss": 0.6313,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.17481562414640808,
|
| 237 |
+
"grad_norm": 2.1510416920205997,
|
| 238 |
+
"learning_rate": 9.832498751495832e-06,
|
| 239 |
+
"loss": 0.637,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.18027861240098333,
|
| 244 |
+
"grad_norm": 2.368980517738678,
|
| 245 |
+
"learning_rate": 9.807125990898905e-06,
|
| 246 |
+
"loss": 0.636,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1857416006555586,
|
| 251 |
+
"grad_norm": 2.2294247264281553,
|
| 252 |
+
"learning_rate": 9.780002124429377e-06,
|
| 253 |
+
"loss": 0.6378,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.19120458891013384,
|
| 258 |
+
"grad_norm": 2.044434482857163,
|
| 259 |
+
"learning_rate": 9.75113703257758e-06,
|
| 260 |
+
"loss": 0.6299,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1966675771647091,
|
| 265 |
+
"grad_norm": 2.273754929604351,
|
| 266 |
+
"learning_rate": 9.720541230115113e-06,
|
| 267 |
+
"loss": 0.6278,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.20213056541928434,
|
| 272 |
+
"grad_norm": 1.9049721088945737,
|
| 273 |
+
"learning_rate": 9.688225862264604e-06,
|
| 274 |
+
"loss": 0.6215,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.2075935536738596,
|
| 279 |
+
"grad_norm": 2.2956911916513936,
|
| 280 |
+
"learning_rate": 9.654202700639805e-06,
|
| 281 |
+
"loss": 0.6114,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.21305654192843484,
|
| 286 |
+
"grad_norm": 2.047625547879147,
|
| 287 |
+
"learning_rate": 9.61848413895751e-06,
|
| 288 |
+
"loss": 0.6272,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.2185195301830101,
|
| 293 |
+
"grad_norm": 2.144882455415688,
|
| 294 |
+
"learning_rate": 9.581083188522862e-06,
|
| 295 |
+
"loss": 0.6146,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.22398251843758535,
|
| 300 |
+
"grad_norm": 1.882314204150727,
|
| 301 |
+
"learning_rate": 9.542013473489683e-06,
|
| 302 |
+
"loss": 0.6115,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.2294455066921606,
|
| 307 |
+
"grad_norm": 2.0782768044314746,
|
| 308 |
+
"learning_rate": 9.501289225897566e-06,
|
| 309 |
+
"loss": 0.5965,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.23490849494673585,
|
| 314 |
+
"grad_norm": 1.9514474259034185,
|
| 315 |
+
"learning_rate": 9.458925280487532e-06,
|
| 316 |
+
"loss": 0.6176,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.2403714832013111,
|
| 321 |
+
"grad_norm": 1.8865949054260884,
|
| 322 |
+
"learning_rate": 9.414937069298125e-06,
|
| 323 |
+
"loss": 0.6018,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.24583447145588638,
|
| 328 |
+
"grad_norm": 2.0683179915172962,
|
| 329 |
+
"learning_rate": 9.369340616043948e-06,
|
| 330 |
+
"loss": 0.6105,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.25129745971046163,
|
| 335 |
+
"grad_norm": 2.004631261000376,
|
| 336 |
+
"learning_rate": 9.322152530278658e-06,
|
| 337 |
+
"loss": 0.5931,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.2567604479650369,
|
| 342 |
+
"grad_norm": 1.9653795566834225,
|
| 343 |
+
"learning_rate": 9.273390001344544e-06,
|
| 344 |
+
"loss": 0.5878,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.26222343621961214,
|
| 349 |
+
"grad_norm": 2.0291055623314715,
|
| 350 |
+
"learning_rate": 9.223070792110927e-06,
|
| 351 |
+
"loss": 0.5827,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2676864244741874,
|
| 356 |
+
"grad_norm": 1.9570477475533896,
|
| 357 |
+
"learning_rate": 9.17121323250362e-06,
|
| 358 |
+
"loss": 0.5943,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.27314941272876264,
|
| 363 |
+
"grad_norm": 2.1357120267635463,
|
| 364 |
+
"learning_rate": 9.117836212827839e-06,
|
| 365 |
+
"loss": 0.571,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2786124009833379,
|
| 370 |
+
"grad_norm": 1.9807416332273953,
|
| 371 |
+
"learning_rate": 9.062959176886967e-06,
|
| 372 |
+
"loss": 0.5895,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.28407538923791315,
|
| 377 |
+
"grad_norm": 1.8679007980966873,
|
| 378 |
+
"learning_rate": 9.006602114899711e-06,
|
| 379 |
+
"loss": 0.5837,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2895383774924884,
|
| 384 |
+
"grad_norm": 2.159508129093504,
|
| 385 |
+
"learning_rate": 8.948785556218202e-06,
|
| 386 |
+
"loss": 0.5872,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.29500136574706365,
|
| 391 |
+
"grad_norm": 2.011039338549636,
|
| 392 |
+
"learning_rate": 8.88953056184971e-06,
|
| 393 |
+
"loss": 0.5656,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.3004643540016389,
|
| 398 |
+
"grad_norm": 1.8497602272423903,
|
| 399 |
+
"learning_rate": 8.828858716784692e-06,
|
| 400 |
+
"loss": 0.5607,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.30592734225621415,
|
| 405 |
+
"grad_norm": 1.8697548121419354,
|
| 406 |
+
"learning_rate": 8.766792122133949e-06,
|
| 407 |
+
"loss": 0.569,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.3113903305107894,
|
| 412 |
+
"grad_norm": 2.023364352460863,
|
| 413 |
+
"learning_rate": 8.703353387077813e-06,
|
| 414 |
+
"loss": 0.5694,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.31685331876536466,
|
| 419 |
+
"grad_norm": 1.9966954532417194,
|
| 420 |
+
"learning_rate": 8.638565620630219e-06,
|
| 421 |
+
"loss": 0.5485,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.3223163070199399,
|
| 426 |
+
"grad_norm": 2.118048147240604,
|
| 427 |
+
"learning_rate": 8.572452423220717e-06,
|
| 428 |
+
"loss": 0.5544,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.32777929527451516,
|
| 433 |
+
"grad_norm": 1.9646992858208574,
|
| 434 |
+
"learning_rate": 8.505037878097481e-06,
|
| 435 |
+
"loss": 0.5585,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3332422835290904,
|
| 440 |
+
"grad_norm": 1.842699727967433,
|
| 441 |
+
"learning_rate": 8.436346542554432e-06,
|
| 442 |
+
"loss": 0.5525,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.33870527178366566,
|
| 447 |
+
"grad_norm": 2.1516397434897,
|
| 448 |
+
"learning_rate": 8.366403438985675e-06,
|
| 449 |
+
"loss": 0.5598,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.3441682600382409,
|
| 454 |
+
"grad_norm": 1.935829862534685,
|
| 455 |
+
"learning_rate": 8.295234045770524e-06,
|
| 456 |
+
"loss": 0.5474,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.34963124829281617,
|
| 461 |
+
"grad_norm": 2.0947580703115,
|
| 462 |
+
"learning_rate": 8.222864287992419e-06,
|
| 463 |
+
"loss": 0.5444,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.3550942365473914,
|
| 468 |
+
"grad_norm": 2.0230909691945462,
|
| 469 |
+
"learning_rate": 8.149320527995111e-06,
|
| 470 |
+
"loss": 0.5367,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.36055722480196667,
|
| 475 |
+
"grad_norm": 1.9422498738110598,
|
| 476 |
+
"learning_rate": 8.074629555779586e-06,
|
| 477 |
+
"loss": 0.5374,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3660202130565419,
|
| 482 |
+
"grad_norm": 2.017157014853384,
|
| 483 |
+
"learning_rate": 7.998818579245183e-06,
|
| 484 |
+
"loss": 0.538,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.3714832013111172,
|
| 489 |
+
"grad_norm": 2.0200129213489952,
|
| 490 |
+
"learning_rate": 7.9219152142785e-06,
|
| 491 |
+
"loss": 0.5261,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.3769461895656924,
|
| 496 |
+
"grad_norm": 1.9025863914724077,
|
| 497 |
+
"learning_rate": 7.843947474693665e-06,
|
| 498 |
+
"loss": 0.5272,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.3824091778202677,
|
| 503 |
+
"grad_norm": 1.98859186872293,
|
| 504 |
+
"learning_rate": 7.764943762027675e-06,
|
| 505 |
+
"loss": 0.5271,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.38787216607484293,
|
| 510 |
+
"grad_norm": 1.9641160396193444,
|
| 511 |
+
"learning_rate": 7.68493285519447e-06,
|
| 512 |
+
"loss": 0.5156,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.3933351543294182,
|
| 517 |
+
"grad_norm": 1.9337347711635438,
|
| 518 |
+
"learning_rate": 7.603943900001567e-06,
|
| 519 |
+
"loss": 0.5152,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.39879814258399343,
|
| 524 |
+
"grad_norm": 1.9734924840604742,
|
| 525 |
+
"learning_rate": 7.522006398533022e-06,
|
| 526 |
+
"loss": 0.5157,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.4042611308385687,
|
| 531 |
+
"grad_norm": 2.3339176407449904,
|
| 532 |
+
"learning_rate": 7.439150198402619e-06,
|
| 533 |
+
"loss": 0.5162,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.40972411909314393,
|
| 538 |
+
"grad_norm": 2.2975439038971732,
|
| 539 |
+
"learning_rate": 7.355405481881207e-06,
|
| 540 |
+
"loss": 0.5157,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.4151871073477192,
|
| 545 |
+
"grad_norm": 1.971313224728527,
|
| 546 |
+
"learning_rate": 7.270802754902092e-06,
|
| 547 |
+
"loss": 0.5014,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.42065009560229444,
|
| 552 |
+
"grad_norm": 1.8926014206142687,
|
| 553 |
+
"learning_rate": 7.185372835948574e-06,
|
| 554 |
+
"loss": 0.5082,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.4261130838568697,
|
| 559 |
+
"grad_norm": 2.0424297512212677,
|
| 560 |
+
"learning_rate": 7.0991468448275914e-06,
|
| 561 |
+
"loss": 0.4969,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.43157607211144494,
|
| 566 |
+
"grad_norm": 1.9075652902136018,
|
| 567 |
+
"learning_rate": 7.0121561913336255e-06,
|
| 568 |
+
"loss": 0.5025,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.4370390603660202,
|
| 573 |
+
"grad_norm": 1.972187929750843,
|
| 574 |
+
"learning_rate": 6.924432563806962e-06,
|
| 575 |
+
"loss": 0.4805,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.44250204862059545,
|
| 580 |
+
"grad_norm": 2.242916059633561,
|
| 581 |
+
"learning_rate": 6.836007917590487e-06,
|
| 582 |
+
"loss": 0.503,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.4479650368751707,
|
| 587 |
+
"grad_norm": 2.0649875803912274,
|
| 588 |
+
"learning_rate": 6.746914463389216e-06,
|
| 589 |
+
"loss": 0.4901,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.45342802512974595,
|
| 594 |
+
"grad_norm": 1.8063638082708948,
|
| 595 |
+
"learning_rate": 6.65718465553681e-06,
|
| 596 |
+
"loss": 0.4755,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.4588910133843212,
|
| 601 |
+
"grad_norm": 1.969494410176558,
|
| 602 |
+
"learning_rate": 6.566851180173344e-06,
|
| 603 |
+
"loss": 0.4874,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.46435400163889645,
|
| 608 |
+
"grad_norm": 1.8970828426082733,
|
| 609 |
+
"learning_rate": 6.475946943338616e-06,
|
| 610 |
+
"loss": 0.4766,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.4698169898934717,
|
| 615 |
+
"grad_norm": 1.9623505460334127,
|
| 616 |
+
"learning_rate": 6.384505058985388e-06,
|
| 617 |
+
"loss": 0.487,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.47527997814804696,
|
| 622 |
+
"grad_norm": 2.016298870850614,
|
| 623 |
+
"learning_rate": 6.292558836916856e-06,
|
| 624 |
+
"loss": 0.4684,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.4807429664026222,
|
| 629 |
+
"grad_norm": 1.9782532484829365,
|
| 630 |
+
"learning_rate": 6.200141770652791e-06,
|
| 631 |
+
"loss": 0.4655,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.4862059546571975,
|
| 636 |
+
"grad_norm": 1.9875880346692325,
|
| 637 |
+
"learning_rate": 6.107287525228764e-06,
|
| 638 |
+
"loss": 0.4829,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.49166894291177277,
|
| 643 |
+
"grad_norm": 2.0320364233161836,
|
| 644 |
+
"learning_rate": 6.014029924932874e-06,
|
| 645 |
+
"loss": 0.4753,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.497131931166348,
|
| 650 |
+
"grad_norm": 2.025675106544592,
|
| 651 |
+
"learning_rate": 5.920402940984483e-06,
|
| 652 |
+
"loss": 0.4673,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.5025949194209233,
|
| 657 |
+
"grad_norm": 1.921954531735398,
|
| 658 |
+
"learning_rate": 5.826440679159424e-06,
|
| 659 |
+
"loss": 0.4532,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.5080579076754985,
|
| 664 |
+
"grad_norm": 2.01066093050374,
|
| 665 |
+
"learning_rate": 5.732177367366176e-06,
|
| 666 |
+
"loss": 0.4658,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.5135208959300738,
|
| 671 |
+
"grad_norm": 1.8422653149296644,
|
| 672 |
+
"learning_rate": 5.63764734317758e-06,
|
| 673 |
+
"loss": 0.4529,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.518983884184649,
|
| 678 |
+
"grad_norm": 1.9255817468742156,
|
| 679 |
+
"learning_rate": 5.542885041322577e-06,
|
| 680 |
+
"loss": 0.4561,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5244468724392243,
|
| 685 |
+
"grad_norm": 1.951977869616497,
|
| 686 |
+
"learning_rate": 5.447924981142578e-06,
|
| 687 |
+
"loss": 0.4565,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.5299098606937995,
|
| 692 |
+
"grad_norm": 1.9554258723769773,
|
| 693 |
+
"learning_rate": 5.3528017540169974e-06,
|
| 694 |
+
"loss": 0.4475,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5353728489483748,
|
| 699 |
+
"grad_norm": 1.9102850022081725,
|
| 700 |
+
"learning_rate": 5.25755001076255e-06,
|
| 701 |
+
"loss": 0.4497,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.54083583720295,
|
| 706 |
+
"grad_norm": 2.239195466962009,
|
| 707 |
+
"learning_rate": 5.1622044490108994e-06,
|
| 708 |
+
"loss": 0.4454,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5462988254575253,
|
| 713 |
+
"grad_norm": 1.8827718246138567,
|
| 714 |
+
"learning_rate": 5.066799800569247e-06,
|
| 715 |
+
"loss": 0.4404,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.5517618137121005,
|
| 720 |
+
"grad_norm": 1.9069058786965245,
|
| 721 |
+
"learning_rate": 4.971370818768475e-06,
|
| 722 |
+
"loss": 0.4236,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.5572248019666758,
|
| 727 |
+
"grad_norm": 1.855279389571921,
|
| 728 |
+
"learning_rate": 4.875952265803452e-06,
|
| 729 |
+
"loss": 0.4403,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.562687790221251,
|
| 734 |
+
"grad_norm": 1.8972542181259846,
|
| 735 |
+
"learning_rate": 4.780578900070104e-06,
|
| 736 |
+
"loss": 0.4372,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5681507784758263,
|
| 741 |
+
"grad_norm": 1.844147614606819,
|
| 742 |
+
"learning_rate": 4.685285463503867e-06,
|
| 743 |
+
"loss": 0.4331,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5736137667304015,
|
| 748 |
+
"grad_norm": 1.9625504078338127,
|
| 749 |
+
"learning_rate": 4.5901066689241514e-06,
|
| 750 |
+
"loss": 0.4292,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5790767549849768,
|
| 755 |
+
"grad_norm": 2.2025005649073433,
|
| 756 |
+
"learning_rate": 4.49507718738939e-06,
|
| 757 |
+
"loss": 0.4293,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.584539743239552,
|
| 762 |
+
"grad_norm": 1.9450789672739341,
|
| 763 |
+
"learning_rate": 4.400231635567319e-06,
|
| 764 |
+
"loss": 0.4192,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5900027314941273,
|
| 769 |
+
"grad_norm": 1.9764711012724976,
|
| 770 |
+
"learning_rate": 4.305604563125061e-06,
|
| 771 |
+
"loss": 0.4182,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5954657197487025,
|
| 776 |
+
"grad_norm": 2.1198793543903194,
|
| 777 |
+
"learning_rate": 4.211230440143604e-06,
|
| 778 |
+
"loss": 0.4114,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.6009287080032778,
|
| 783 |
+
"grad_norm": 3.1276187198521463,
|
| 784 |
+
"learning_rate": 4.117143644561306e-06,
|
| 785 |
+
"loss": 0.4166,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.606391696257853,
|
| 790 |
+
"grad_norm": 1.8794840574959752,
|
| 791 |
+
"learning_rate": 4.023378449650928e-06,
|
| 792 |
+
"loss": 0.4076,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.6118546845124283,
|
| 797 |
+
"grad_norm": 1.9139438651443652,
|
| 798 |
+
"learning_rate": 3.929969011534811e-06,
|
| 799 |
+
"loss": 0.4218,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.6173176727670036,
|
| 804 |
+
"grad_norm": 1.861741261098825,
|
| 805 |
+
"learning_rate": 3.83694935674272e-06,
|
| 806 |
+
"loss": 0.4169,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.6227806610215788,
|
| 811 |
+
"grad_norm": 1.8247668127301933,
|
| 812 |
+
"learning_rate": 3.7443533698169186e-06,
|
| 813 |
+
"loss": 0.4132,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.6282436492761541,
|
| 818 |
+
"grad_norm": 1.9796553518989775,
|
| 819 |
+
"learning_rate": 3.652214780968926e-06,
|
| 820 |
+
"loss": 0.3996,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.6337066375307293,
|
| 825 |
+
"grad_norm": 1.823241119094308,
|
| 826 |
+
"learning_rate": 3.5605671537925264e-06,
|
| 827 |
+
"loss": 0.4088,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.6391696257853046,
|
| 832 |
+
"grad_norm": 1.881819370184891,
|
| 833 |
+
"learning_rate": 3.469443873037457e-06,
|
| 834 |
+
"loss": 0.4073,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.6446326140398798,
|
| 839 |
+
"grad_norm": 1.947962800229108,
|
| 840 |
+
"learning_rate": 3.3788781324482446e-06,
|
| 841 |
+
"loss": 0.4067,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6500956022944551,
|
| 846 |
+
"grad_norm": 1.9214089256240032,
|
| 847 |
+
"learning_rate": 3.2889029226726286e-06,
|
| 848 |
+
"loss": 0.3996,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6555585905490303,
|
| 853 |
+
"grad_norm": 1.9543022070758702,
|
| 854 |
+
"learning_rate": 3.1995510192439587e-06,
|
| 855 |
+
"loss": 0.3982,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6610215788036056,
|
| 860 |
+
"grad_norm": 1.9173316158851936,
|
| 861 |
+
"learning_rate": 3.1108549706419553e-06,
|
| 862 |
+
"loss": 0.4,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.6664845670581808,
|
| 867 |
+
"grad_norm": 1.8159793606682484,
|
| 868 |
+
"learning_rate": 3.022847086436176e-06,
|
| 869 |
+
"loss": 0.3988,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.6719475553127561,
|
| 874 |
+
"grad_norm": 1.9089073637989584,
|
| 875 |
+
"learning_rate": 2.9355594255165186e-06,
|
| 876 |
+
"loss": 0.3871,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.6774105435673313,
|
| 881 |
+
"grad_norm": 1.8914866485656172,
|
| 882 |
+
"learning_rate": 2.8490237844150335e-06,
|
| 883 |
+
"loss": 0.3885,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.6828735318219066,
|
| 888 |
+
"grad_norm": 1.9221955744591699,
|
| 889 |
+
"learning_rate": 2.763271685723311e-06,
|
| 890 |
+
"loss": 0.3818,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.6883365200764818,
|
| 895 |
+
"grad_norm": 1.8733166396673193,
|
| 896 |
+
"learning_rate": 2.6783343666096445e-06,
|
| 897 |
+
"loss": 0.3874,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.6937995083310571,
|
| 902 |
+
"grad_norm": 2.134783287581667,
|
| 903 |
+
"learning_rate": 2.5942427674401936e-06,
|
| 904 |
+
"loss": 0.3899,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6992624965856323,
|
| 909 |
+
"grad_norm": 1.8202198608461309,
|
| 910 |
+
"learning_rate": 2.5110275205082224e-06,
|
| 911 |
+
"loss": 0.3839,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.7047254848402076,
|
| 916 |
+
"grad_norm": 1.9005906531792638,
|
| 917 |
+
"learning_rate": 2.428718938875607e-06,
|
| 918 |
+
"loss": 0.3812,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.7101884730947828,
|
| 923 |
+
"grad_norm": 1.8656177124830933,
|
| 924 |
+
"learning_rate": 2.3473470053305954e-06,
|
| 925 |
+
"loss": 0.3805,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.7156514613493581,
|
| 930 |
+
"grad_norm": 1.9350473980336609,
|
| 931 |
+
"learning_rate": 2.266941361465886e-06,
|
| 932 |
+
"loss": 0.381,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.7211144496039333,
|
| 937 |
+
"grad_norm": 1.8146852411399312,
|
| 938 |
+
"learning_rate": 2.187531296881017e-06,
|
| 939 |
+
"loss": 0.3751,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.7265774378585086,
|
| 944 |
+
"grad_norm": 1.8246829733561232,
|
| 945 |
+
"learning_rate": 2.1091457385129267e-06,
|
| 946 |
+
"loss": 0.3752,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.7320404261130838,
|
| 951 |
+
"grad_norm": 1.8550222595976351,
|
| 952 |
+
"learning_rate": 2.031813240098686e-06,
|
| 953 |
+
"loss": 0.3721,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7375034143676591,
|
| 958 |
+
"grad_norm": 1.8354415599143141,
|
| 959 |
+
"learning_rate": 1.9555619717741248e-06,
|
| 960 |
+
"loss": 0.3733,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7429664026222343,
|
| 965 |
+
"grad_norm": 1.9427997534753285,
|
| 966 |
+
"learning_rate": 1.8804197098122168e-06,
|
| 967 |
+
"loss": 0.3745,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.7484293908768096,
|
| 972 |
+
"grad_norm": 1.9154466908271706,
|
| 973 |
+
"learning_rate": 1.8064138265049457e-06,
|
| 974 |
+
"loss": 0.3743,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.7538923791313848,
|
| 979 |
+
"grad_norm": 1.8117953280714802,
|
| 980 |
+
"learning_rate": 1.7335712801923015e-06,
|
| 981 |
+
"loss": 0.3707,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.7593553673859601,
|
| 986 |
+
"grad_norm": 1.894844068057027,
|
| 987 |
+
"learning_rate": 1.6619186054421088e-06,
|
| 988 |
+
"loss": 0.3677,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.7648183556405354,
|
| 993 |
+
"grad_norm": 1.8949515530814003,
|
| 994 |
+
"learning_rate": 1.5914819033841843e-06,
|
| 995 |
+
"loss": 0.3639,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.7702813438951106,
|
| 1000 |
+
"grad_norm": 1.7568307263853036,
|
| 1001 |
+
"learning_rate": 1.5222868322024092e-06,
|
| 1002 |
+
"loss": 0.3586,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.7757443321496859,
|
| 1007 |
+
"grad_norm": 1.759178792643852,
|
| 1008 |
+
"learning_rate": 1.4543585977881513e-06,
|
| 1009 |
+
"loss": 0.35,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.7812073204042611,
|
| 1014 |
+
"grad_norm": 1.8973466785062572,
|
| 1015 |
+
"learning_rate": 1.3877219445584283e-06,
|
| 1016 |
+
"loss": 0.3593,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.7866703086588364,
|
| 1021 |
+
"grad_norm": 1.9605293056514521,
|
| 1022 |
+
"learning_rate": 1.3224011464422032e-06,
|
| 1023 |
+
"loss": 0.3621,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.7921332969134116,
|
| 1028 |
+
"grad_norm": 1.812660380797101,
|
| 1029 |
+
"learning_rate": 1.2584199980380358e-06,
|
| 1030 |
+
"loss": 0.3599,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.7975962851679869,
|
| 1035 |
+
"grad_norm": 1.8356228496720945,
|
| 1036 |
+
"learning_rate": 1.1958018059463577e-06,
|
| 1037 |
+
"loss": 0.3605,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.8030592734225621,
|
| 1042 |
+
"grad_norm": 1.691309969533953,
|
| 1043 |
+
"learning_rate": 1.1345693802795177e-06,
|
| 1044 |
+
"loss": 0.3583,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.8085222616771374,
|
| 1049 |
+
"grad_norm": 1.8242600919765277,
|
| 1050 |
+
"learning_rate": 1.0747450263526576e-06,
|
| 1051 |
+
"loss": 0.3598,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.8139852499317126,
|
| 1056 |
+
"grad_norm": 1.7765876082507979,
|
| 1057 |
+
"learning_rate": 1.0163505365585086e-06,
|
| 1058 |
+
"loss": 0.3493,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.8194482381862879,
|
| 1063 |
+
"grad_norm": 1.8095739075310635,
|
| 1064 |
+
"learning_rate": 9.594071824289986e-07,
|
| 1065 |
+
"loss": 0.3571,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.8249112264408631,
|
| 1070 |
+
"grad_norm": 1.7811537133839481,
|
| 1071 |
+
"learning_rate": 9.039357068866178e-07,
|
| 1072 |
+
"loss": 0.3518,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.8303742146954384,
|
| 1077 |
+
"grad_norm": 1.7625847558197685,
|
| 1078 |
+
"learning_rate": 8.49956316688329e-07,
|
| 1079 |
+
"loss": 0.3518,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.8358372029500136,
|
| 1084 |
+
"grad_norm": 1.8189872840713601,
|
| 1085 |
+
"learning_rate": 7.974886750647887e-07,
|
| 1086 |
+
"loss": 0.3381,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.8413001912045889,
|
| 1091 |
+
"grad_norm": 1.8024026803759359,
|
| 1092 |
+
"learning_rate": 7.465518945575789e-07,
|
| 1093 |
+
"loss": 0.345,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.8467631794591641,
|
| 1098 |
+
"grad_norm": 1.895342687896103,
|
| 1099 |
+
"learning_rate": 6.971645300570084e-07,
|
| 1100 |
+
"loss": 0.3464,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.8522261677137394,
|
| 1105 |
+
"grad_norm": 1.796319045444646,
|
| 1106 |
+
"learning_rate": 6.49344572043083e-07,
|
| 1107 |
+
"loss": 0.3477,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.8576891559683146,
|
| 1112 |
+
"grad_norm": 1.8845046027721593,
|
| 1113 |
+
"learning_rate": 6.031094400320498e-07,
|
| 1114 |
+
"loss": 0.3443,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.8631521442228899,
|
| 1119 |
+
"grad_norm": 1.724637234062916,
|
| 1120 |
+
"learning_rate": 5.584759762309422e-07,
|
| 1121 |
+
"loss": 0.3367,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8686151324774651,
|
| 1126 |
+
"grad_norm": 1.8964510379949302,
|
| 1127 |
+
"learning_rate": 5.154604394024254e-07,
|
| 1128 |
+
"loss": 0.3396,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.8740781207320404,
|
| 1133 |
+
"grad_norm": 1.6763296815328177,
|
| 1134 |
+
"learning_rate": 4.7407849894216097e-07,
|
| 1135 |
+
"loss": 0.3348,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.8795411089866156,
|
| 1140 |
+
"grad_norm": 1.7128909727748394,
|
| 1141 |
+
"learning_rate": 4.343452291708783e-07,
|
| 1142 |
+
"loss": 0.343,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.8850040972411909,
|
| 1147 |
+
"grad_norm": 1.8153223119786095,
|
| 1148 |
+
"learning_rate": 3.962751038431961e-07,
|
| 1149 |
+
"loss": 0.3405,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.8904670854957661,
|
| 1154 |
+
"grad_norm": 1.7625733448640701,
|
| 1155 |
+
"learning_rate": 3.5988199087523987e-07,
|
| 1156 |
+
"loss": 0.3396,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.8959300737503414,
|
| 1161 |
+
"grad_norm": 1.8570474171994868,
|
| 1162 |
+
"learning_rate": 3.2517914729292446e-07,
|
| 1163 |
+
"loss": 0.3391,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.9013930620049166,
|
| 1168 |
+
"grad_norm": 1.8466215461741358,
|
| 1169 |
+
"learning_rate": 2.921792144027813e-07,
|
| 1170 |
+
"loss": 0.3407,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.9068560502594919,
|
| 1175 |
+
"grad_norm": 1.7424828950437026,
|
| 1176 |
+
"learning_rate": 2.6089421318707785e-07,
|
| 1177 |
+
"loss": 0.3365,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.9123190385140671,
|
| 1182 |
+
"grad_norm": 1.8348358804511151,
|
| 1183 |
+
"learning_rate": 2.3133553992488845e-07,
|
| 1184 |
+
"loss": 0.3453,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.9177820267686424,
|
| 1189 |
+
"grad_norm": 1.8267323618432934,
|
| 1190 |
+
"learning_rate": 2.0351396204074946e-07,
|
| 1191 |
+
"loss": 0.337,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.9232450150232177,
|
| 1196 |
+
"grad_norm": 1.6632279811187767,
|
| 1197 |
+
"learning_rate": 1.7743961418237144e-07,
|
| 1198 |
+
"loss": 0.3341,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.9287080032777929,
|
| 1203 |
+
"grad_norm": 1.8291284979726596,
|
| 1204 |
+
"learning_rate": 1.5312199452886145e-07,
|
| 1205 |
+
"loss": 0.332,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.9341709915323682,
|
| 1210 |
+
"grad_norm": 1.7047344118755032,
|
| 1211 |
+
"learning_rate": 1.3056996133079924e-07,
|
| 1212 |
+
"loss": 0.3375,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.9396339797869434,
|
| 1217 |
+
"grad_norm": 1.745785425026002,
|
| 1218 |
+
"learning_rate": 1.0979172968340667e-07,
|
| 1219 |
+
"loss": 0.3321,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.9450969680415187,
|
| 1224 |
+
"grad_norm": 1.8776860033466447,
|
| 1225 |
+
"learning_rate": 9.079486853402097e-08,
|
| 1226 |
+
"loss": 0.34,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.9505599562960939,
|
| 1231 |
+
"grad_norm": 1.7269678812008018,
|
| 1232 |
+
"learning_rate": 7.358629792492522e-08,
|
| 1233 |
+
"loss": 0.342,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.9560229445506692,
|
| 1238 |
+
"grad_norm": 1.7727161663929172,
|
| 1239 |
+
"learning_rate": 5.8172286472566456e-08,
|
| 1240 |
+
"loss": 0.3391,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.9614859328052444,
|
| 1245 |
+
"grad_norm": 1.7673439074550763,
|
| 1246 |
+
"learning_rate": 4.455844908407059e-08,
|
| 1247 |
+
"loss": 0.3416,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.9669489210598197,
|
| 1252 |
+
"grad_norm": 1.7746641186916818,
|
| 1253 |
+
"learning_rate": 3.27497449118791e-08,
|
| 1254 |
+
"loss": 0.3268,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.972411909314395,
|
| 1259 |
+
"grad_norm": 1.858186055296109,
|
| 1260 |
+
"learning_rate": 2.275047554726795e-08,
|
| 1261 |
+
"loss": 0.3371,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.9778748975689703,
|
| 1266 |
+
"grad_norm": 1.6930712261364957,
|
| 1267 |
+
"learning_rate": 1.4564283453392669e-08,
|
| 1268 |
+
"loss": 0.3425,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.9833378858235455,
|
| 1273 |
+
"grad_norm": 1.778838215839912,
|
| 1274 |
+
"learning_rate": 8.194150638438093e-09,
|
| 1275 |
+
"loss": 0.3315,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.9888008740781208,
|
| 1280 |
+
"grad_norm": 1.8180837953030582,
|
| 1281 |
+
"learning_rate": 3.6423975693533618e-09,
|
| 1282 |
+
"loss": 0.3404,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.994263862332696,
|
| 1287 |
+
"grad_norm": 1.7628508134053706,
|
| 1288 |
+
"learning_rate": 9.106823265681019e-10,
|
| 1289 |
+
"loss": 0.335,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.9997268505872713,
|
| 1294 |
+
"grad_norm": 1.7810702627412893,
|
| 1295 |
+
"learning_rate": 0.0,
|
| 1296 |
+
"loss": 0.3383,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.9997268505872713,
|
| 1301 |
+
"eval_loss": 0.2970966100692749,
|
| 1302 |
+
"eval_runtime": 1.1854,
|
| 1303 |
+
"eval_samples_per_second": 1.687,
|
| 1304 |
+
"eval_steps_per_second": 0.844,
|
| 1305 |
+
"step": 915
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 0.9997268505872713,
|
| 1309 |
+
"step": 915,
|
| 1310 |
+
"total_flos": 191530040033280.0,
|
| 1311 |
+
"train_loss": 0.4996221627042593,
|
| 1312 |
+
"train_runtime": 20787.7504,
|
| 1313 |
+
"train_samples_per_second": 1.409,
|
| 1314 |
+
"train_steps_per_second": 0.044
|
| 1315 |
+
}
|
| 1316 |
+
],
|
| 1317 |
+
"logging_steps": 5,
|
| 1318 |
+
"max_steps": 915,
|
| 1319 |
+
"num_input_tokens_seen": 0,
|
| 1320 |
+
"num_train_epochs": 1,
|
| 1321 |
+
"save_steps": 100,
|
| 1322 |
+
"stateful_callbacks": {
|
| 1323 |
+
"TrainerControl": {
|
| 1324 |
+
"args": {
|
| 1325 |
+
"should_epoch_stop": false,
|
| 1326 |
+
"should_evaluate": false,
|
| 1327 |
+
"should_log": false,
|
| 1328 |
+
"should_save": true,
|
| 1329 |
+
"should_training_stop": true
|
| 1330 |
+
},
|
| 1331 |
+
"attributes": {}
|
| 1332 |
+
}
|
| 1333 |
+
},
|
| 1334 |
+
"total_flos": 191530040033280.0,
|
| 1335 |
+
"train_batch_size": 2,
|
| 1336 |
+
"trial_name": null,
|
| 1337 |
+
"trial_params": null
|
| 1338 |
+
}
|