Model save
Browse files- README.md +62 -0
- all_results.json +9 -0
- generation_config.json +6 -0
- model.safetensors +1 -1
- runs/Apr21_16-33-54_ai-pc-2/events.out.tfevents.1713692070.ai-pc-2.3645834.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1374 -0
README.md
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
base_model: stabilityai/stablelm-2-1_6b
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: stablelm-2-1_6b-sft-full
|
| 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 |
+
# stablelm-2-1_6b-sft-full
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [stabilityai/stablelm-2-1_6b](https://huggingface.co/stabilityai/stablelm-2-1_6b) on the generator dataset.
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 2e-05
|
| 40 |
+
- train_batch_size: 4
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- distributed_type: multi-GPU
|
| 44 |
+
- num_devices: 2
|
| 45 |
+
- gradient_accumulation_steps: 16
|
| 46 |
+
- total_train_batch_size: 128
|
| 47 |
+
- total_eval_batch_size: 16
|
| 48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 49 |
+
- lr_scheduler_type: cosine
|
| 50 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 51 |
+
- num_epochs: 1
|
| 52 |
+
|
| 53 |
+
### Training results
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
|
| 57 |
+
### Framework versions
|
| 58 |
+
|
| 59 |
+
- Transformers 4.40.0
|
| 60 |
+
- Pytorch 2.1.2
|
| 61 |
+
- Datasets 2.18.0
|
| 62 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9990846083431411,
|
| 3 |
+
"total_flos": 522931729858560.0,
|
| 4 |
+
"train_loss": 1.2670168572071334,
|
| 5 |
+
"train_runtime": 18760.2941,
|
| 6 |
+
"train_samples": 207864,
|
| 7 |
+
"train_samples_per_second": 6.522,
|
| 8 |
+
"train_steps_per_second": 0.051
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 100257,
|
| 4 |
+
"eos_token_id": 100257,
|
| 5 |
+
"transformers_version": "4.40.0"
|
| 6 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3289069520
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49b4815304b09a960cfc66775c350a4320ae429259d0331419ebd66f7de5a004
|
| 3 |
size 3289069520
|
runs/Apr21_16-33-54_ai-pc-2/events.out.tfevents.1713692070.ai-pc-2.3645834.0
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 sha256:1cf376d35fc4dc49d76c7e8db283e0e073720bf7d440deb2f94a3888f02da361
|
| 3 |
+
size 45705
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9990846083431411,
|
| 3 |
+
"total_flos": 522931729858560.0,
|
| 4 |
+
"train_loss": 1.2670168572071334,
|
| 5 |
+
"train_runtime": 18760.2941,
|
| 6 |
+
"train_samples": 207864,
|
| 7 |
+
"train_samples_per_second": 6.522,
|
| 8 |
+
"train_steps_per_second": 0.051
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1374 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9990846083431411,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 955,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0010461618935530272,
|
| 13 |
+
"grad_norm": 4.961991027903084,
|
| 14 |
+
"learning_rate": 2.0833333333333333e-07,
|
| 15 |
+
"loss": 1.5879,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005230809467765137,
|
| 20 |
+
"grad_norm": 4.618544468117898,
|
| 21 |
+
"learning_rate": 1.0416666666666667e-06,
|
| 22 |
+
"loss": 1.5933,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.010461618935530274,
|
| 27 |
+
"grad_norm": 4.037023933749454,
|
| 28 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 29 |
+
"loss": 1.5702,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01569242840329541,
|
| 34 |
+
"grad_norm": 1.8708269586813249,
|
| 35 |
+
"learning_rate": 3.125e-06,
|
| 36 |
+
"loss": 1.5369,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.020923237871060547,
|
| 41 |
+
"grad_norm": 1.5902570000850822,
|
| 42 |
+
"learning_rate": 4.166666666666667e-06,
|
| 43 |
+
"loss": 1.4646,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.026154047338825683,
|
| 48 |
+
"grad_norm": 1.23920395738266,
|
| 49 |
+
"learning_rate": 5.208333333333334e-06,
|
| 50 |
+
"loss": 1.4335,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.03138485680659082,
|
| 55 |
+
"grad_norm": 1.1663956891784453,
|
| 56 |
+
"learning_rate": 6.25e-06,
|
| 57 |
+
"loss": 1.3982,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03661566627435596,
|
| 62 |
+
"grad_norm": 0.8974420486494152,
|
| 63 |
+
"learning_rate": 7.291666666666667e-06,
|
| 64 |
+
"loss": 1.3919,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.041846475742121095,
|
| 69 |
+
"grad_norm": 0.8445948684549006,
|
| 70 |
+
"learning_rate": 8.333333333333334e-06,
|
| 71 |
+
"loss": 1.3576,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.04707728520988623,
|
| 76 |
+
"grad_norm": 0.6980197067702758,
|
| 77 |
+
"learning_rate": 9.375000000000001e-06,
|
| 78 |
+
"loss": 1.3582,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.052308094677651365,
|
| 83 |
+
"grad_norm": 0.8409122561540898,
|
| 84 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 85 |
+
"loss": 1.3458,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.0575389041454165,
|
| 90 |
+
"grad_norm": 0.6948599150905024,
|
| 91 |
+
"learning_rate": 1.1458333333333333e-05,
|
| 92 |
+
"loss": 1.335,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.06276971361318164,
|
| 97 |
+
"grad_norm": 0.651340813779911,
|
| 98 |
+
"learning_rate": 1.25e-05,
|
| 99 |
+
"loss": 1.322,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.06800052308094677,
|
| 104 |
+
"grad_norm": 0.684798803653097,
|
| 105 |
+
"learning_rate": 1.3541666666666668e-05,
|
| 106 |
+
"loss": 1.3456,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.07323133254871192,
|
| 111 |
+
"grad_norm": 0.6258722254574212,
|
| 112 |
+
"learning_rate": 1.4583333333333333e-05,
|
| 113 |
+
"loss": 1.317,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.07846214201647705,
|
| 118 |
+
"grad_norm": 0.661994715425898,
|
| 119 |
+
"learning_rate": 1.5625e-05,
|
| 120 |
+
"loss": 1.314,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08369295148424219,
|
| 125 |
+
"grad_norm": 0.6157440945511082,
|
| 126 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 127 |
+
"loss": 1.3174,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.08892376095200732,
|
| 132 |
+
"grad_norm": 0.6136457513656507,
|
| 133 |
+
"learning_rate": 1.7708333333333335e-05,
|
| 134 |
+
"loss": 1.3208,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.09415457041977246,
|
| 139 |
+
"grad_norm": 0.6609752901635255,
|
| 140 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 141 |
+
"loss": 1.3217,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0993853798875376,
|
| 146 |
+
"grad_norm": 0.6433338679997594,
|
| 147 |
+
"learning_rate": 1.979166666666667e-05,
|
| 148 |
+
"loss": 1.3209,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.10461618935530273,
|
| 153 |
+
"grad_norm": 0.748785028585811,
|
| 154 |
+
"learning_rate": 1.999892997072575e-05,
|
| 155 |
+
"loss": 1.3015,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.10984699882306787,
|
| 160 |
+
"grad_norm": 0.7348845509310983,
|
| 161 |
+
"learning_rate": 1.99945833692589e-05,
|
| 162 |
+
"loss": 1.3043,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.115077808290833,
|
| 167 |
+
"grad_norm": 0.7150167557908745,
|
| 168 |
+
"learning_rate": 1.9986894771071707e-05,
|
| 169 |
+
"loss": 1.3036,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.12030861775859815,
|
| 174 |
+
"grad_norm": 0.6229579295007591,
|
| 175 |
+
"learning_rate": 1.9975866747083734e-05,
|
| 176 |
+
"loss": 1.3046,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.12553942722636327,
|
| 181 |
+
"grad_norm": 0.626634171036282,
|
| 182 |
+
"learning_rate": 1.9961502984854394e-05,
|
| 183 |
+
"loss": 1.3011,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.1307702366941284,
|
| 188 |
+
"grad_norm": 0.6732748047847544,
|
| 189 |
+
"learning_rate": 1.9943808287349902e-05,
|
| 190 |
+
"loss": 1.2922,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.13600104616189354,
|
| 195 |
+
"grad_norm": 0.6336745846509011,
|
| 196 |
+
"learning_rate": 1.992278857133726e-05,
|
| 197 |
+
"loss": 1.3023,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.1412318556296587,
|
| 202 |
+
"grad_norm": 0.63287698051173,
|
| 203 |
+
"learning_rate": 1.9898450865405786e-05,
|
| 204 |
+
"loss": 1.2912,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.14646266509742384,
|
| 209 |
+
"grad_norm": 0.6243758841550209,
|
| 210 |
+
"learning_rate": 1.9870803307616916e-05,
|
| 211 |
+
"loss": 1.2953,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.15169347456518897,
|
| 216 |
+
"grad_norm": 0.7127519633361736,
|
| 217 |
+
"learning_rate": 1.983985514278296e-05,
|
| 218 |
+
"loss": 1.2955,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.1569242840329541,
|
| 223 |
+
"grad_norm": 0.6226333486523956,
|
| 224 |
+
"learning_rate": 1.9805616719375852e-05,
|
| 225 |
+
"loss": 1.2932,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.16215509350071924,
|
| 230 |
+
"grad_norm": 0.6823517100118245,
|
| 231 |
+
"learning_rate": 1.9768099486066776e-05,
|
| 232 |
+
"loss": 1.2745,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.16738590296848438,
|
| 237 |
+
"grad_norm": 0.6663881488848158,
|
| 238 |
+
"learning_rate": 1.9727315987897993e-05,
|
| 239 |
+
"loss": 1.3024,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.17261671243624951,
|
| 244 |
+
"grad_norm": 0.6250204764928626,
|
| 245 |
+
"learning_rate": 1.9683279862087986e-05,
|
| 246 |
+
"loss": 1.2725,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.17784752190401465,
|
| 251 |
+
"grad_norm": 0.6587024144174084,
|
| 252 |
+
"learning_rate": 1.963600583347147e-05,
|
| 253 |
+
"loss": 1.2883,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.18307833137177978,
|
| 258 |
+
"grad_norm": 0.6553453461637575,
|
| 259 |
+
"learning_rate": 1.9585509709575646e-05,
|
| 260 |
+
"loss": 1.2934,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.18830914083954492,
|
| 265 |
+
"grad_norm": 0.6612913681156035,
|
| 266 |
+
"learning_rate": 1.9531808375334512e-05,
|
| 267 |
+
"loss": 1.2663,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.19353995030731005,
|
| 272 |
+
"grad_norm": 0.6536121257096528,
|
| 273 |
+
"learning_rate": 1.9474919787442835e-05,
|
| 274 |
+
"loss": 1.2588,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1987707597750752,
|
| 279 |
+
"grad_norm": 0.6498560967867221,
|
| 280 |
+
"learning_rate": 1.9414862968351788e-05,
|
| 281 |
+
"loss": 1.261,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.20400156924284032,
|
| 286 |
+
"grad_norm": 0.6437028941150709,
|
| 287 |
+
"learning_rate": 1.935165799990821e-05,
|
| 288 |
+
"loss": 1.2764,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.20923237871060546,
|
| 293 |
+
"grad_norm": 0.7042353774035648,
|
| 294 |
+
"learning_rate": 1.9285326016639624e-05,
|
| 295 |
+
"loss": 1.2715,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.2144631881783706,
|
| 300 |
+
"grad_norm": 0.6615769962934966,
|
| 301 |
+
"learning_rate": 1.9215889198687245e-05,
|
| 302 |
+
"loss": 1.2515,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.21969399764613573,
|
| 307 |
+
"grad_norm": 0.6674880822729611,
|
| 308 |
+
"learning_rate": 1.9143370764389374e-05,
|
| 309 |
+
"loss": 1.2828,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.22492480711390087,
|
| 314 |
+
"grad_norm": 0.6423252942504364,
|
| 315 |
+
"learning_rate": 1.906779496251763e-05,
|
| 316 |
+
"loss": 1.2791,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.230155616581666,
|
| 321 |
+
"grad_norm": 0.6391984125238104,
|
| 322 |
+
"learning_rate": 1.8989187064168643e-05,
|
| 323 |
+
"loss": 1.2519,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.23538642604943116,
|
| 328 |
+
"grad_norm": 0.6477151176504692,
|
| 329 |
+
"learning_rate": 1.8907573354313853e-05,
|
| 330 |
+
"loss": 1.2777,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.2406172355171963,
|
| 335 |
+
"grad_norm": 0.6269146787303694,
|
| 336 |
+
"learning_rate": 1.8822981123010343e-05,
|
| 337 |
+
"loss": 1.2736,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.24584804498496143,
|
| 342 |
+
"grad_norm": 0.6477481280746402,
|
| 343 |
+
"learning_rate": 1.873543865627556e-05,
|
| 344 |
+
"loss": 1.2732,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.25107885445272654,
|
| 349 |
+
"grad_norm": 0.6651047763900065,
|
| 350 |
+
"learning_rate": 1.8644975226629025e-05,
|
| 351 |
+
"loss": 1.2829,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.2563096639204917,
|
| 356 |
+
"grad_norm": 0.6432975261843621,
|
| 357 |
+
"learning_rate": 1.8551621083304147e-05,
|
| 358 |
+
"loss": 1.2615,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.2615404733882568,
|
| 363 |
+
"grad_norm": 0.6589800378988344,
|
| 364 |
+
"learning_rate": 1.8455407442133467e-05,
|
| 365 |
+
"loss": 1.2824,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.266771282856022,
|
| 370 |
+
"grad_norm": 0.6374519780745838,
|
| 371 |
+
"learning_rate": 1.8356366475110697e-05,
|
| 372 |
+
"loss": 1.2809,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2720020923237871,
|
| 377 |
+
"grad_norm": 0.6474176000311428,
|
| 378 |
+
"learning_rate": 1.8254531299633007e-05,
|
| 379 |
+
"loss": 1.2765,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.27723290179155224,
|
| 384 |
+
"grad_norm": 0.611205619408698,
|
| 385 |
+
"learning_rate": 1.81499359674272e-05,
|
| 386 |
+
"loss": 1.2723,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.2824637112593174,
|
| 391 |
+
"grad_norm": 0.680714525048176,
|
| 392 |
+
"learning_rate": 1.8042615453163484e-05,
|
| 393 |
+
"loss": 1.2606,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2876945207270825,
|
| 398 |
+
"grad_norm": 0.6405723682781137,
|
| 399 |
+
"learning_rate": 1.7932605642760607e-05,
|
| 400 |
+
"loss": 1.2495,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2929253301948477,
|
| 405 |
+
"grad_norm": 0.6279226914713609,
|
| 406 |
+
"learning_rate": 1.7819943321386295e-05,
|
| 407 |
+
"loss": 1.2735,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2981561396626128,
|
| 412 |
+
"grad_norm": 0.6284127404605249,
|
| 413 |
+
"learning_rate": 1.7704666161156994e-05,
|
| 414 |
+
"loss": 1.241,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.30338694913037795,
|
| 419 |
+
"grad_norm": 0.6499789986037162,
|
| 420 |
+
"learning_rate": 1.7586812708541046e-05,
|
| 421 |
+
"loss": 1.2607,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.30861775859814305,
|
| 426 |
+
"grad_norm": 0.6557734964553777,
|
| 427 |
+
"learning_rate": 1.746642237146948e-05,
|
| 428 |
+
"loss": 1.2746,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.3138485680659082,
|
| 433 |
+
"grad_norm": 0.6397583630089397,
|
| 434 |
+
"learning_rate": 1.7343535406158773e-05,
|
| 435 |
+
"loss": 1.2606,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3190793775336733,
|
| 440 |
+
"grad_norm": 0.6893845618635109,
|
| 441 |
+
"learning_rate": 1.7218192903649926e-05,
|
| 442 |
+
"loss": 1.261,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.3243101870014385,
|
| 447 |
+
"grad_norm": 0.7201666359135361,
|
| 448 |
+
"learning_rate": 1.7090436776068422e-05,
|
| 449 |
+
"loss": 1.2713,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.3295409964692036,
|
| 454 |
+
"grad_norm": 0.6687526125254434,
|
| 455 |
+
"learning_rate": 1.6960309742609603e-05,
|
| 456 |
+
"loss": 1.2721,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.33477180593696876,
|
| 461 |
+
"grad_norm": 0.7677992466340372,
|
| 462 |
+
"learning_rate": 1.682785531525422e-05,
|
| 463 |
+
"loss": 1.2671,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.34000261540473387,
|
| 468 |
+
"grad_norm": 0.7054430863360542,
|
| 469 |
+
"learning_rate": 1.6693117784218818e-05,
|
| 470 |
+
"loss": 1.2512,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.34523342487249903,
|
| 475 |
+
"grad_norm": 0.7122244372275465,
|
| 476 |
+
"learning_rate": 1.655614220314598e-05,
|
| 477 |
+
"loss": 1.2625,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.35046423434026414,
|
| 482 |
+
"grad_norm": 0.679853303463346,
|
| 483 |
+
"learning_rate": 1.6416974374039227e-05,
|
| 484 |
+
"loss": 1.2767,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.3556950438080293,
|
| 489 |
+
"grad_norm": 0.6684546083293239,
|
| 490 |
+
"learning_rate": 1.6275660831947725e-05,
|
| 491 |
+
"loss": 1.253,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.3609258532757944,
|
| 496 |
+
"grad_norm": 0.6148491341875125,
|
| 497 |
+
"learning_rate": 1.6132248829405845e-05,
|
| 498 |
+
"loss": 1.2664,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.36615666274355957,
|
| 503 |
+
"grad_norm": 0.6287400163765243,
|
| 504 |
+
"learning_rate": 1.5986786320632842e-05,
|
| 505 |
+
"loss": 1.29,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3713874722113247,
|
| 510 |
+
"grad_norm": 0.6429840333148633,
|
| 511 |
+
"learning_rate": 1.5839321945497847e-05,
|
| 512 |
+
"loss": 1.259,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.37661828167908984,
|
| 517 |
+
"grad_norm": 0.6213894267805022,
|
| 518 |
+
"learning_rate": 1.5689905013255683e-05,
|
| 519 |
+
"loss": 1.2721,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.381849091146855,
|
| 524 |
+
"grad_norm": 0.6657140337203303,
|
| 525 |
+
"learning_rate": 1.5538585486058747e-05,
|
| 526 |
+
"loss": 1.2629,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3870799006146201,
|
| 531 |
+
"grad_norm": 0.6216486922374719,
|
| 532 |
+
"learning_rate": 1.5385413962250657e-05,
|
| 533 |
+
"loss": 1.2445,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.39231071008238527,
|
| 538 |
+
"grad_norm": 0.640435105766003,
|
| 539 |
+
"learning_rate": 1.5230441659447128e-05,
|
| 540 |
+
"loss": 1.2672,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3975415195501504,
|
| 545 |
+
"grad_norm": 0.628008495827186,
|
| 546 |
+
"learning_rate": 1.507372039740978e-05,
|
| 547 |
+
"loss": 1.2782,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.40277232901791554,
|
| 552 |
+
"grad_norm": 0.6525051028721182,
|
| 553 |
+
"learning_rate": 1.4915302580718614e-05,
|
| 554 |
+
"loss": 1.249,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.40800313848568065,
|
| 559 |
+
"grad_norm": 0.6278258712485207,
|
| 560 |
+
"learning_rate": 1.4755241181248923e-05,
|
| 561 |
+
"loss": 1.2588,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.4132339479534458,
|
| 566 |
+
"grad_norm": 0.6550288396521164,
|
| 567 |
+
"learning_rate": 1.4593589720458507e-05,
|
| 568 |
+
"loss": 1.255,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.4184647574212109,
|
| 573 |
+
"grad_norm": 0.6820552488018682,
|
| 574 |
+
"learning_rate": 1.443040225149114e-05,
|
| 575 |
+
"loss": 1.2498,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.4236955668889761,
|
| 580 |
+
"grad_norm": 0.6399254481819071,
|
| 581 |
+
"learning_rate": 1.4265733341102235e-05,
|
| 582 |
+
"loss": 1.2666,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.4289263763567412,
|
| 587 |
+
"grad_norm": 0.6722624199607894,
|
| 588 |
+
"learning_rate": 1.4099638051412745e-05,
|
| 589 |
+
"loss": 1.245,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.43415718582450635,
|
| 594 |
+
"grad_norm": 0.6302145600803009,
|
| 595 |
+
"learning_rate": 1.3932171921497483e-05,
|
| 596 |
+
"loss": 1.2611,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.43938799529227146,
|
| 601 |
+
"grad_norm": 0.6031404765855133,
|
| 602 |
+
"learning_rate": 1.3763390948813897e-05,
|
| 603 |
+
"loss": 1.2372,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.4446188047600366,
|
| 608 |
+
"grad_norm": 0.646493114663968,
|
| 609 |
+
"learning_rate": 1.3593351570477608e-05,
|
| 610 |
+
"loss": 1.2426,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.44984961422780173,
|
| 615 |
+
"grad_norm": 0.6195059553793902,
|
| 616 |
+
"learning_rate": 1.3422110644390911e-05,
|
| 617 |
+
"loss": 1.2169,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.4550804236955669,
|
| 622 |
+
"grad_norm": 0.6012958936997463,
|
| 623 |
+
"learning_rate": 1.3249725430230595e-05,
|
| 624 |
+
"loss": 1.2728,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.460311233163332,
|
| 629 |
+
"grad_norm": 0.617177322353044,
|
| 630 |
+
"learning_rate": 1.3076253570301409e-05,
|
| 631 |
+
"loss": 1.2518,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.46554204263109716,
|
| 636 |
+
"grad_norm": 0.6292828261202444,
|
| 637 |
+
"learning_rate": 1.2901753070261565e-05,
|
| 638 |
+
"loss": 1.2543,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.4707728520988623,
|
| 643 |
+
"grad_norm": 0.597666355379697,
|
| 644 |
+
"learning_rate": 1.2726282279726788e-05,
|
| 645 |
+
"loss": 1.2427,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.47600366156662743,
|
| 650 |
+
"grad_norm": 0.6234161905971162,
|
| 651 |
+
"learning_rate": 1.2549899872759288e-05,
|
| 652 |
+
"loss": 1.2602,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.4812344710343926,
|
| 657 |
+
"grad_norm": 0.629762327137135,
|
| 658 |
+
"learning_rate": 1.237266482824832e-05,
|
| 659 |
+
"loss": 1.2428,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.4864652805021577,
|
| 664 |
+
"grad_norm": 0.6210086267634197,
|
| 665 |
+
"learning_rate": 1.2194636410188748e-05,
|
| 666 |
+
"loss": 1.2281,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.49169608996992287,
|
| 671 |
+
"grad_norm": 0.6456911580326817,
|
| 672 |
+
"learning_rate": 1.2015874147864314e-05,
|
| 673 |
+
"loss": 1.2342,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.496926899437688,
|
| 678 |
+
"grad_norm": 0.6472528263693088,
|
| 679 |
+
"learning_rate": 1.183643781594219e-05,
|
| 680 |
+
"loss": 1.2427,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5021577089054531,
|
| 685 |
+
"grad_norm": 0.647823531177134,
|
| 686 |
+
"learning_rate": 1.165638741448548e-05,
|
| 687 |
+
"loss": 1.2522,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.5073885183732183,
|
| 692 |
+
"grad_norm": 0.6693118299281153,
|
| 693 |
+
"learning_rate": 1.147578314889033e-05,
|
| 694 |
+
"loss": 1.2619,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5126193278409834,
|
| 699 |
+
"grad_norm": 0.667730694230485,
|
| 700 |
+
"learning_rate": 1.1294685409754434e-05,
|
| 701 |
+
"loss": 1.2429,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.5178501373087485,
|
| 706 |
+
"grad_norm": 0.6046644310580812,
|
| 707 |
+
"learning_rate": 1.1113154752683548e-05,
|
| 708 |
+
"loss": 1.2588,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5230809467765136,
|
| 713 |
+
"grad_norm": 0.6785327532515957,
|
| 714 |
+
"learning_rate": 1.0931251878042882e-05,
|
| 715 |
+
"loss": 1.2673,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.5283117562442788,
|
| 720 |
+
"grad_norm": 0.6252703595295017,
|
| 721 |
+
"learning_rate": 1.0749037610660041e-05,
|
| 722 |
+
"loss": 1.264,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.533542565712044,
|
| 727 |
+
"grad_norm": 0.6079580441878377,
|
| 728 |
+
"learning_rate": 1.0566572879486388e-05,
|
| 729 |
+
"loss": 1.2619,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.538773375179809,
|
| 734 |
+
"grad_norm": 0.6213036767388472,
|
| 735 |
+
"learning_rate": 1.0383918697223564e-05,
|
| 736 |
+
"loss": 1.2602,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5440041846475742,
|
| 741 |
+
"grad_norm": 0.5992653066437161,
|
| 742 |
+
"learning_rate": 1.020113613992203e-05,
|
| 743 |
+
"loss": 1.2636,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5492349941153394,
|
| 748 |
+
"grad_norm": 0.5983825011403769,
|
| 749 |
+
"learning_rate": 1.001828632655837e-05,
|
| 750 |
+
"loss": 1.254,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.5544658035831045,
|
| 755 |
+
"grad_norm": 0.6237947953829833,
|
| 756 |
+
"learning_rate": 9.835430398598319e-06,
|
| 757 |
+
"loss": 1.2573,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.5596966130508696,
|
| 762 |
+
"grad_norm": 0.6015544745764206,
|
| 763 |
+
"learning_rate": 9.652629499552216e-06,
|
| 764 |
+
"loss": 1.2332,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5649274225186348,
|
| 769 |
+
"grad_norm": 0.6724219499123357,
|
| 770 |
+
"learning_rate": 9.469944754529784e-06,
|
| 771 |
+
"loss": 1.2489,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5701582319863999,
|
| 776 |
+
"grad_norm": 0.5900962511293482,
|
| 777 |
+
"learning_rate": 9.28743724980107e-06,
|
| 778 |
+
"loss": 1.2486,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.575389041454165,
|
| 783 |
+
"grad_norm": 0.5939356137294083,
|
| 784 |
+
"learning_rate": 9.105168012370372e-06,
|
| 785 |
+
"loss": 1.2464,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5806198509219301,
|
| 790 |
+
"grad_norm": 0.6072005219993765,
|
| 791 |
+
"learning_rate": 8.923197989569981e-06,
|
| 792 |
+
"loss": 1.2484,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.5858506603896954,
|
| 797 |
+
"grad_norm": 0.6408535640602984,
|
| 798 |
+
"learning_rate": 8.741588028680566e-06,
|
| 799 |
+
"loss": 1.2577,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.5910814698574605,
|
| 804 |
+
"grad_norm": 0.5820294502690248,
|
| 805 |
+
"learning_rate": 8.560398856585002e-06,
|
| 806 |
+
"loss": 1.2519,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.5963122793252256,
|
| 811 |
+
"grad_norm": 0.5866823280811903,
|
| 812 |
+
"learning_rate": 8.379691059462478e-06,
|
| 813 |
+
"loss": 1.2373,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.6015430887929907,
|
| 818 |
+
"grad_norm": 0.6268376105145648,
|
| 819 |
+
"learning_rate": 8.199525062529626e-06,
|
| 820 |
+
"loss": 1.261,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.6067738982607559,
|
| 825 |
+
"grad_norm": 0.5899507751154045,
|
| 826 |
+
"learning_rate": 8.01996110983552e-06,
|
| 827 |
+
"loss": 1.2469,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.612004707728521,
|
| 832 |
+
"grad_norm": 0.6110501827689294,
|
| 833 |
+
"learning_rate": 7.841059244117189e-06,
|
| 834 |
+
"loss": 1.24,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.6172355171962861,
|
| 839 |
+
"grad_norm": 0.628795675438309,
|
| 840 |
+
"learning_rate": 7.662879286722496e-06,
|
| 841 |
+
"loss": 1.2678,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6224663266640512,
|
| 846 |
+
"grad_norm": 0.5817062103966367,
|
| 847 |
+
"learning_rate": 7.485480817607031e-06,
|
| 848 |
+
"loss": 1.2384,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6276971361318164,
|
| 853 |
+
"grad_norm": 0.5950726911519293,
|
| 854 |
+
"learning_rate": 7.30892315541171e-06,
|
| 855 |
+
"loss": 1.2356,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6329279455995815,
|
| 860 |
+
"grad_norm": 0.5727952261916434,
|
| 861 |
+
"learning_rate": 7.133265337627757e-06,
|
| 862 |
+
"loss": 1.2406,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.6381587550673467,
|
| 867 |
+
"grad_norm": 0.575271616339552,
|
| 868 |
+
"learning_rate": 6.958566100855716e-06,
|
| 869 |
+
"loss": 1.2437,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.6433895645351118,
|
| 874 |
+
"grad_norm": 0.5790390123089401,
|
| 875 |
+
"learning_rate": 6.78488386116505e-06,
|
| 876 |
+
"loss": 1.2583,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.648620374002877,
|
| 881 |
+
"grad_norm": 0.5932077205106131,
|
| 882 |
+
"learning_rate": 6.612276694560927e-06,
|
| 883 |
+
"loss": 1.2413,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.6538511834706421,
|
| 888 |
+
"grad_norm": 0.5947522152199447,
|
| 889 |
+
"learning_rate": 6.44080231756473e-06,
|
| 890 |
+
"loss": 1.2549,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.6590819929384072,
|
| 895 |
+
"grad_norm": 1.3459652842247485,
|
| 896 |
+
"learning_rate": 6.2705180679147455e-06,
|
| 897 |
+
"loss": 1.2387,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.6643128024061724,
|
| 902 |
+
"grad_norm": 0.7269750930986757,
|
| 903 |
+
"learning_rate": 6.101480885393537e-06,
|
| 904 |
+
"loss": 1.2315,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6695436118739375,
|
| 909 |
+
"grad_norm": 0.5957998707048122,
|
| 910 |
+
"learning_rate": 5.933747292788369e-06,
|
| 911 |
+
"loss": 1.2677,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.6747744213417026,
|
| 916 |
+
"grad_norm": 0.5850290378596165,
|
| 917 |
+
"learning_rate": 5.767373376991082e-06,
|
| 918 |
+
"loss": 1.2291,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6800052308094677,
|
| 923 |
+
"grad_norm": 0.5844027863277863,
|
| 924 |
+
"learning_rate": 5.602414770243698e-06,
|
| 925 |
+
"loss": 1.2273,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.685236040277233,
|
| 930 |
+
"grad_norm": 0.5868679965624024,
|
| 931 |
+
"learning_rate": 5.438926631536087e-06,
|
| 932 |
+
"loss": 1.2479,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.6904668497449981,
|
| 937 |
+
"grad_norm": 0.5831449921364225,
|
| 938 |
+
"learning_rate": 5.276963628161833e-06,
|
| 939 |
+
"loss": 1.247,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.6956976592127632,
|
| 944 |
+
"grad_norm": 0.5755759622405301,
|
| 945 |
+
"learning_rate": 5.116579917438564e-06,
|
| 946 |
+
"loss": 1.2447,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.7009284686805283,
|
| 951 |
+
"grad_norm": 0.5768540427743302,
|
| 952 |
+
"learning_rate": 4.957829128598781e-06,
|
| 953 |
+
"loss": 1.2177,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7061592781482935,
|
| 958 |
+
"grad_norm": 0.598911774868325,
|
| 959 |
+
"learning_rate": 4.80076434485727e-06,
|
| 960 |
+
"loss": 1.2321,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7113900876160586,
|
| 965 |
+
"grad_norm": 0.5969706523304447,
|
| 966 |
+
"learning_rate": 4.645438085661085e-06,
|
| 967 |
+
"loss": 1.234,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.7166208970838237,
|
| 972 |
+
"grad_norm": 0.5866245773773069,
|
| 973 |
+
"learning_rate": 4.4919022891280725e-06,
|
| 974 |
+
"loss": 1.2456,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.7218517065515888,
|
| 979 |
+
"grad_norm": 0.5811720592986388,
|
| 980 |
+
"learning_rate": 4.340208294679745e-06,
|
| 981 |
+
"loss": 1.245,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.727082516019354,
|
| 986 |
+
"grad_norm": 0.5849016337714122,
|
| 987 |
+
"learning_rate": 4.190406825874377e-06,
|
| 988 |
+
"loss": 1.2212,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.7323133254871191,
|
| 993 |
+
"grad_norm": 0.6255138841124418,
|
| 994 |
+
"learning_rate": 4.042547973446017e-06,
|
| 995 |
+
"loss": 1.2318,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.7375441349548842,
|
| 1000 |
+
"grad_norm": 0.5808819882391798,
|
| 1001 |
+
"learning_rate": 3.896681178555099e-06,
|
| 1002 |
+
"loss": 1.2616,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.7427749444226494,
|
| 1007 |
+
"grad_norm": 0.5821049889121402,
|
| 1008 |
+
"learning_rate": 3.7528552162562858e-06,
|
| 1009 |
+
"loss": 1.2342,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.7480057538904146,
|
| 1014 |
+
"grad_norm": 0.6008223090596293,
|
| 1015 |
+
"learning_rate": 3.6111181791890184e-06,
|
| 1016 |
+
"loss": 1.2472,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.7532365633581797,
|
| 1021 |
+
"grad_norm": 0.5950331976607752,
|
| 1022 |
+
"learning_rate": 3.471517461496253e-06,
|
| 1023 |
+
"loss": 1.2211,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.7584673728259448,
|
| 1028 |
+
"grad_norm": 0.5695566788307966,
|
| 1029 |
+
"learning_rate": 3.3340997429767786e-06,
|
| 1030 |
+
"loss": 1.2397,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.76369818229371,
|
| 1035 |
+
"grad_norm": 0.5922108634556766,
|
| 1036 |
+
"learning_rate": 3.1989109734763936e-06,
|
| 1037 |
+
"loss": 1.2624,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.7689289917614751,
|
| 1042 |
+
"grad_norm": 0.573161649411567,
|
| 1043 |
+
"learning_rate": 3.0659963575231544e-06,
|
| 1044 |
+
"loss": 1.25,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.7741598012292402,
|
| 1049 |
+
"grad_norm": 0.5745082815289888,
|
| 1050 |
+
"learning_rate": 2.935400339211841e-06,
|
| 1051 |
+
"loss": 1.2564,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.7793906106970053,
|
| 1056 |
+
"grad_norm": 0.5599070185795609,
|
| 1057 |
+
"learning_rate": 2.8071665873427244e-06,
|
| 1058 |
+
"loss": 1.2341,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7846214201647705,
|
| 1063 |
+
"grad_norm": 0.5798846902278265,
|
| 1064 |
+
"learning_rate": 2.681337980819536e-06,
|
| 1065 |
+
"loss": 1.2425,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7898522296325357,
|
| 1070 |
+
"grad_norm": 0.5630366958265783,
|
| 1071 |
+
"learning_rate": 2.5579565943116092e-06,
|
| 1072 |
+
"loss": 1.2503,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.7950830391003008,
|
| 1077 |
+
"grad_norm": 0.5612114479196867,
|
| 1078 |
+
"learning_rate": 2.437063684184893e-06,
|
| 1079 |
+
"loss": 1.2372,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.8003138485680659,
|
| 1084 |
+
"grad_norm": 0.5786482688071404,
|
| 1085 |
+
"learning_rate": 2.318699674706639e-06,
|
| 1086 |
+
"loss": 1.2539,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.8055446580358311,
|
| 1091 |
+
"grad_norm": 0.5609594448423239,
|
| 1092 |
+
"learning_rate": 2.202904144528295e-06,
|
| 1093 |
+
"loss": 1.2359,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.8107754675035962,
|
| 1098 |
+
"grad_norm": 0.5761081087482597,
|
| 1099 |
+
"learning_rate": 2.08971581345115e-06,
|
| 1100 |
+
"loss": 1.2333,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.8160062769713613,
|
| 1105 |
+
"grad_norm": 0.5773902237533778,
|
| 1106 |
+
"learning_rate": 1.979172529479193e-06,
|
| 1107 |
+
"loss": 1.2405,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.8212370864391264,
|
| 1112 |
+
"grad_norm": 0.5692272061773305,
|
| 1113 |
+
"learning_rate": 1.8713112561634671e-06,
|
| 1114 |
+
"loss": 1.2531,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.8264678959068916,
|
| 1119 |
+
"grad_norm": 0.8502635478696626,
|
| 1120 |
+
"learning_rate": 1.7661680602421594e-06,
|
| 1121 |
+
"loss": 1.2357,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8316987053746567,
|
| 1126 |
+
"grad_norm": 0.5725480612758482,
|
| 1127 |
+
"learning_rate": 1.663778099580583e-06,
|
| 1128 |
+
"loss": 1.231,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.8369295148424218,
|
| 1133 |
+
"grad_norm": 0.5528630417115946,
|
| 1134 |
+
"learning_rate": 1.5641756114150552e-06,
|
| 1135 |
+
"loss": 1.2412,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.842160324310187,
|
| 1140 |
+
"grad_norm": 0.5462565230167501,
|
| 1141 |
+
"learning_rate": 1.4673939009046268e-06,
|
| 1142 |
+
"loss": 1.2233,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.8473911337779522,
|
| 1147 |
+
"grad_norm": 0.5663269502388393,
|
| 1148 |
+
"learning_rate": 1.3734653299944834e-06,
|
| 1149 |
+
"loss": 1.2277,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.8526219432457173,
|
| 1154 |
+
"grad_norm": 0.5686962762287001,
|
| 1155 |
+
"learning_rate": 1.2824213065947232e-06,
|
| 1156 |
+
"loss": 1.2357,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.8578527527134824,
|
| 1161 |
+
"grad_norm": 0.5638758372127466,
|
| 1162 |
+
"learning_rate": 1.194292274078156e-06,
|
| 1163 |
+
"loss": 1.2377,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.8630835621812476,
|
| 1168 |
+
"grad_norm": 0.5586378869453262,
|
| 1169 |
+
"learning_rate": 1.1091077011006302e-06,
|
| 1170 |
+
"loss": 1.2223,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.8683143716490127,
|
| 1175 |
+
"grad_norm": 0.5517524656609795,
|
| 1176 |
+
"learning_rate": 1.0268960717472742e-06,
|
| 1177 |
+
"loss": 1.2574,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.8735451811167778,
|
| 1182 |
+
"grad_norm": 0.5569662486368889,
|
| 1183 |
+
"learning_rate": 9.476848760079671e-07,
|
| 1184 |
+
"loss": 1.2335,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.8787759905845429,
|
| 1189 |
+
"grad_norm": 0.5643447607687533,
|
| 1190 |
+
"learning_rate": 8.715006005852144e-07,
|
| 1191 |
+
"loss": 1.2304,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.8840068000523081,
|
| 1196 |
+
"grad_norm": 0.550747548292025,
|
| 1197 |
+
"learning_rate": 7.983687200375046e-07,
|
| 1198 |
+
"loss": 1.2281,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.8892376095200732,
|
| 1203 |
+
"grad_norm": 0.5574658998449357,
|
| 1204 |
+
"learning_rate": 7.283136882611063e-07,
|
| 1205 |
+
"loss": 1.2329,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.8944684189878384,
|
| 1210 |
+
"grad_norm": 0.5737161178123568,
|
| 1211 |
+
"learning_rate": 6.613589303131506e-07,
|
| 1212 |
+
"loss": 1.2465,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.8996992284556035,
|
| 1217 |
+
"grad_norm": 0.5738100002167115,
|
| 1218 |
+
"learning_rate": 5.975268345787455e-07,
|
| 1219 |
+
"loss": 1.2408,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.9049300379233687,
|
| 1224 |
+
"grad_norm": 0.5542103705854088,
|
| 1225 |
+
"learning_rate": 5.368387452847312e-07,
|
| 1226 |
+
"loss": 1.2369,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.9101608473911338,
|
| 1231 |
+
"grad_norm": 0.5509438118396119,
|
| 1232 |
+
"learning_rate": 4.793149553625786e-07,
|
| 1233 |
+
"loss": 1.2677,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.9153916568588989,
|
| 1238 |
+
"grad_norm": 0.5573824119903792,
|
| 1239 |
+
"learning_rate": 4.2497469966282125e-07,
|
| 1240 |
+
"loss": 1.2429,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.920622466326664,
|
| 1245 |
+
"grad_norm": 0.5438647452016272,
|
| 1246 |
+
"learning_rate": 3.738361485232922e-07,
|
| 1247 |
+
"loss": 1.2236,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.9258532757944292,
|
| 1252 |
+
"grad_norm": 0.5597553842668248,
|
| 1253 |
+
"learning_rate": 3.2591640169331697e-07,
|
| 1254 |
+
"loss": 1.2443,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.9310840852621943,
|
| 1259 |
+
"grad_norm": 0.5541351488188309,
|
| 1260 |
+
"learning_rate": 2.8123148261587465e-07,
|
| 1261 |
+
"loss": 1.2292,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.9363148947299594,
|
| 1266 |
+
"grad_norm": 0.5510390666909667,
|
| 1267 |
+
"learning_rate": 2.397963330696751e-07,
|
| 1268 |
+
"loss": 1.2371,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.9415457041977247,
|
| 1273 |
+
"grad_norm": 0.5662372990839701,
|
| 1274 |
+
"learning_rate": 2.0162480817291442e-07,
|
| 1275 |
+
"loss": 1.2385,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.9467765136654898,
|
| 1280 |
+
"grad_norm": 0.5633092432947229,
|
| 1281 |
+
"learning_rate": 1.6672967175038634e-07,
|
| 1282 |
+
"loss": 1.2357,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.9520073231332549,
|
| 1287 |
+
"grad_norm": 0.5485802540619047,
|
| 1288 |
+
"learning_rate": 1.3512259206550748e-07,
|
| 1289 |
+
"loss": 1.241,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.95723813260102,
|
| 1294 |
+
"grad_norm": 0.5451187400528298,
|
| 1295 |
+
"learning_rate": 1.0681413791867157e-07,
|
| 1296 |
+
"loss": 1.2331,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.9624689420687852,
|
| 1301 |
+
"grad_norm": 0.5544164676125732,
|
| 1302 |
+
"learning_rate": 8.181377511324306e-08,
|
| 1303 |
+
"loss": 1.2486,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.9676997515365503,
|
| 1308 |
+
"grad_norm": 0.5773845105032677,
|
| 1309 |
+
"learning_rate": 6.012986329038462e-08,
|
| 1310 |
+
"loss": 1.2386,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.9729305610043154,
|
| 1315 |
+
"grad_norm": 0.5478843295079262,
|
| 1316 |
+
"learning_rate": 4.1769653133743036e-08,
|
| 1317 |
+
"loss": 1.215,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.9781613704720805,
|
| 1322 |
+
"grad_norm": 0.5608458529460807,
|
| 1323 |
+
"learning_rate": 2.673928394496206e-08,
|
| 1324 |
+
"loss": 1.2301,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.9833921799398457,
|
| 1329 |
+
"grad_norm": 0.5519105435517161,
|
| 1330 |
+
"learning_rate": 1.5043781590823313e-08,
|
| 1331 |
+
"loss": 1.2173,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.9886229894076108,
|
| 1336 |
+
"grad_norm": 0.5536445777152418,
|
| 1337 |
+
"learning_rate": 6.687056822688442e-09,
|
| 1338 |
+
"loss": 1.2308,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.993853798875376,
|
| 1343 |
+
"grad_norm": 0.5686300705360647,
|
| 1344 |
+
"learning_rate": 1.6719039688162242e-09,
|
| 1345 |
+
"loss": 1.2192,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.9990846083431411,
|
| 1350 |
+
"grad_norm": 0.5671702033849003,
|
| 1351 |
+
"learning_rate": 0.0,
|
| 1352 |
+
"loss": 1.24,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.9990846083431411,
|
| 1357 |
+
"step": 955,
|
| 1358 |
+
"total_flos": 522931729858560.0,
|
| 1359 |
+
"train_loss": 1.2670168572071334,
|
| 1360 |
+
"train_runtime": 18760.2941,
|
| 1361 |
+
"train_samples_per_second": 6.522,
|
| 1362 |
+
"train_steps_per_second": 0.051
|
| 1363 |
+
}
|
| 1364 |
+
],
|
| 1365 |
+
"logging_steps": 5,
|
| 1366 |
+
"max_steps": 955,
|
| 1367 |
+
"num_input_tokens_seen": 0,
|
| 1368 |
+
"num_train_epochs": 1,
|
| 1369 |
+
"save_steps": 100,
|
| 1370 |
+
"total_flos": 522931729858560.0,
|
| 1371 |
+
"train_batch_size": 4,
|
| 1372 |
+
"trial_name": null,
|
| 1373 |
+
"trial_params": null
|
| 1374 |
+
}
|