Model save
Browse files- README.md +69 -0
- adapter_model.safetensors +1 -1
- all_results.json +9 -0
- runs/Sep29_15-26-06_5bc1778c8f32/events.out.tfevents.1727623620.5bc1778c8f32.108129.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1730 -0
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: google/gemma-7b
|
| 3 |
+
datasets:
|
| 4 |
+
- generator
|
| 5 |
+
library_name: peft
|
| 6 |
+
license: gemma
|
| 7 |
+
tags:
|
| 8 |
+
- trl
|
| 9 |
+
- sft
|
| 10 |
+
- generated_from_trainer
|
| 11 |
+
model-index:
|
| 12 |
+
- name: gemma7b-gpt4o_100k_coding-lora
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# gemma7b-gpt4o_100k_coding-lora
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.2800
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 0.0003
|
| 43 |
+
- train_batch_size: 1
|
| 44 |
+
- eval_batch_size: 1
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 8
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 16
|
| 50 |
+
- total_eval_batch_size: 8
|
| 51 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
+
- lr_scheduler_type: cosine
|
| 53 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 54 |
+
- num_epochs: 1
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 60 |
+
| 0.5901 | 0.9996 | 1195 | 1.2800 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- PEFT 0.13.0
|
| 66 |
+
- Transformers 4.45.1
|
| 67 |
+
- Pytorch 2.4.1+cu121
|
| 68 |
+
- Datasets 3.0.1
|
| 69 |
+
- Tokenizers 0.20.0
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 12859872
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:39878badd29124b975bb8df5bfaf7a1627e690d82b3797f228aa8ea1241bf1f7
|
| 3 |
size 12859872
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.999581764951903,
|
| 3 |
+
"total_flos": 9.109418934146171e+17,
|
| 4 |
+
"train_loss": 1.440422640086218,
|
| 5 |
+
"train_runtime": 6570.4384,
|
| 6 |
+
"train_samples": 116368,
|
| 7 |
+
"train_samples_per_second": 2.911,
|
| 8 |
+
"train_steps_per_second": 0.182
|
| 9 |
+
}
|
runs/Sep29_15-26-06_5bc1778c8f32/events.out.tfevents.1727623620.5bc1778c8f32.108129.0
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:fd12868a08c79bd23240846e1ea4c9c082f8de4033a5a7cbd2eb6e428aa3f56e
|
| 3 |
+
size 57855
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.999581764951903,
|
| 3 |
+
"total_flos": 9.109418934146171e+17,
|
| 4 |
+
"train_loss": 1.440422640086218,
|
| 5 |
+
"train_runtime": 6570.4384,
|
| 6 |
+
"train_samples": 116368,
|
| 7 |
+
"train_samples_per_second": 2.911,
|
| 8 |
+
"train_steps_per_second": 0.182
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1730 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.999581764951903,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1195,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.000836470096194061,
|
| 13 |
+
"grad_norm": 109.33856201171875,
|
| 14 |
+
"learning_rate": 2.4999999999999998e-06,
|
| 15 |
+
"loss": 24.0328,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004182350480970306,
|
| 20 |
+
"grad_norm": 97.91401672363281,
|
| 21 |
+
"learning_rate": 1.2499999999999999e-05,
|
| 22 |
+
"loss": 22.4374,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.008364700961940611,
|
| 27 |
+
"grad_norm": 36.984405517578125,
|
| 28 |
+
"learning_rate": 2.4999999999999998e-05,
|
| 29 |
+
"loss": 20.9539,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.012547051442910916,
|
| 34 |
+
"grad_norm": 20.562854766845703,
|
| 35 |
+
"learning_rate": 3.75e-05,
|
| 36 |
+
"loss": 18.0157,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.016729401923881223,
|
| 41 |
+
"grad_norm": 7.916600227355957,
|
| 42 |
+
"learning_rate": 4.9999999999999996e-05,
|
| 43 |
+
"loss": 15.4241,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.020911752404851526,
|
| 48 |
+
"grad_norm": 6.871030807495117,
|
| 49 |
+
"learning_rate": 6.25e-05,
|
| 50 |
+
"loss": 14.6159,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.025094102885821833,
|
| 55 |
+
"grad_norm": 6.986784934997559,
|
| 56 |
+
"learning_rate": 7.5e-05,
|
| 57 |
+
"loss": 13.6058,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.029276453366792136,
|
| 62 |
+
"grad_norm": 3.074172019958496,
|
| 63 |
+
"learning_rate": 8.75e-05,
|
| 64 |
+
"loss": 12.7264,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.033458803847762446,
|
| 69 |
+
"grad_norm": 2.548049211502075,
|
| 70 |
+
"learning_rate": 9.999999999999999e-05,
|
| 71 |
+
"loss": 12.7985,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.037641154328732745,
|
| 76 |
+
"grad_norm": 3.184255361557007,
|
| 77 |
+
"learning_rate": 0.0001125,
|
| 78 |
+
"loss": 12.2921,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04182350480970305,
|
| 83 |
+
"grad_norm": 4.834357738494873,
|
| 84 |
+
"learning_rate": 0.000125,
|
| 85 |
+
"loss": 11.7192,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04600585529067336,
|
| 90 |
+
"grad_norm": 8.06936264038086,
|
| 91 |
+
"learning_rate": 0.00013749999999999998,
|
| 92 |
+
"loss": 11.0815,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.050188205771643665,
|
| 97 |
+
"grad_norm": 12.874434471130371,
|
| 98 |
+
"learning_rate": 0.00015,
|
| 99 |
+
"loss": 9.8728,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05437055625261397,
|
| 104 |
+
"grad_norm": 18.794525146484375,
|
| 105 |
+
"learning_rate": 0.00016249999999999997,
|
| 106 |
+
"loss": 7.7488,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05855290673358427,
|
| 111 |
+
"grad_norm": 21.42744255065918,
|
| 112 |
+
"learning_rate": 0.000175,
|
| 113 |
+
"loss": 4.8002,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.06273525721455459,
|
| 118 |
+
"grad_norm": 7.021483898162842,
|
| 119 |
+
"learning_rate": 0.00018749999999999998,
|
| 120 |
+
"loss": 2.2667,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06691760769552489,
|
| 125 |
+
"grad_norm": 4.35729455947876,
|
| 126 |
+
"learning_rate": 0.00019999999999999998,
|
| 127 |
+
"loss": 1.7236,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.07109995817649518,
|
| 132 |
+
"grad_norm": 2.531404972076416,
|
| 133 |
+
"learning_rate": 0.0002125,
|
| 134 |
+
"loss": 1.4388,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07528230865746549,
|
| 139 |
+
"grad_norm": 1.7126120328903198,
|
| 140 |
+
"learning_rate": 0.000225,
|
| 141 |
+
"loss": 1.2603,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.0794646591384358,
|
| 146 |
+
"grad_norm": 1.713827133178711,
|
| 147 |
+
"learning_rate": 0.00023749999999999997,
|
| 148 |
+
"loss": 1.1149,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.0836470096194061,
|
| 153 |
+
"grad_norm": 0.6418541073799133,
|
| 154 |
+
"learning_rate": 0.00025,
|
| 155 |
+
"loss": 1.0645,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08782936010037641,
|
| 160 |
+
"grad_norm": 0.9687772989273071,
|
| 161 |
+
"learning_rate": 0.0002625,
|
| 162 |
+
"loss": 1.0047,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.09201171058134672,
|
| 167 |
+
"grad_norm": 0.8204954266548157,
|
| 168 |
+
"learning_rate": 0.00027499999999999996,
|
| 169 |
+
"loss": 0.986,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09619406106231702,
|
| 174 |
+
"grad_norm": 0.5046463012695312,
|
| 175 |
+
"learning_rate": 0.0002875,
|
| 176 |
+
"loss": 0.9229,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.10037641154328733,
|
| 181 |
+
"grad_norm": 1.1709442138671875,
|
| 182 |
+
"learning_rate": 0.0003,
|
| 183 |
+
"loss": 0.9344,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.10455876202425764,
|
| 188 |
+
"grad_norm": 0.9831328392028809,
|
| 189 |
+
"learning_rate": 0.0002999839868651235,
|
| 190 |
+
"loss": 0.8812,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.10874111250522794,
|
| 195 |
+
"grad_norm": 1.4019944667816162,
|
| 196 |
+
"learning_rate": 0.0002999359508794339,
|
| 197 |
+
"loss": 0.8814,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.11292346298619825,
|
| 202 |
+
"grad_norm": 0.78233802318573,
|
| 203 |
+
"learning_rate": 0.00029985590229902073,
|
| 204 |
+
"loss": 0.8701,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.11710581346716854,
|
| 209 |
+
"grad_norm": 1.4517076015472412,
|
| 210 |
+
"learning_rate": 0.0002997438582149335,
|
| 211 |
+
"loss": 0.8753,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.12128816394813885,
|
| 216 |
+
"grad_norm": 2.320331335067749,
|
| 217 |
+
"learning_rate": 0.0002995998425495327,
|
| 218 |
+
"loss": 0.8464,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.12547051442910917,
|
| 223 |
+
"grad_norm": 1.0486135482788086,
|
| 224 |
+
"learning_rate": 0.000299423886051382,
|
| 225 |
+
"loss": 0.8498,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.12965286491007946,
|
| 230 |
+
"grad_norm": 0.9131810069084167,
|
| 231 |
+
"learning_rate": 0.0002992160262886831,
|
| 232 |
+
"loss": 0.8468,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.13383521539104978,
|
| 237 |
+
"grad_norm": 1.4284504652023315,
|
| 238 |
+
"learning_rate": 0.0002989763076412549,
|
| 239 |
+
"loss": 0.8088,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.13801756587202008,
|
| 244 |
+
"grad_norm": 0.6106438040733337,
|
| 245 |
+
"learning_rate": 0.000298704781291058,
|
| 246 |
+
"loss": 0.8215,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.14219991635299037,
|
| 251 |
+
"grad_norm": 0.5358127951622009,
|
| 252 |
+
"learning_rate": 0.0002984015052112665,
|
| 253 |
+
"loss": 0.8201,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1463822668339607,
|
| 258 |
+
"grad_norm": 1.5017443895339966,
|
| 259 |
+
"learning_rate": 0.0002980665441538907,
|
| 260 |
+
"loss": 0.7957,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.15056461731493098,
|
| 265 |
+
"grad_norm": 1.1214942932128906,
|
| 266 |
+
"learning_rate": 0.00029769996963595184,
|
| 267 |
+
"loss": 0.8083,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1547469677959013,
|
| 272 |
+
"grad_norm": 2.1036202907562256,
|
| 273 |
+
"learning_rate": 0.0002973018599242125,
|
| 274 |
+
"loss": 0.7929,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.1589293182768716,
|
| 279 |
+
"grad_norm": 1.0557819604873657,
|
| 280 |
+
"learning_rate": 0.0002968723000184662,
|
| 281 |
+
"loss": 0.7868,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.16311166875784192,
|
| 286 |
+
"grad_norm": 0.9558168649673462,
|
| 287 |
+
"learning_rate": 0.00029641138163338907,
|
| 288 |
+
"loss": 0.7812,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.1672940192388122,
|
| 293 |
+
"grad_norm": 0.771851122379303,
|
| 294 |
+
"learning_rate": 0.0002959192031789579,
|
| 295 |
+
"loss": 0.7846,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.17147636971978253,
|
| 300 |
+
"grad_norm": 1.288865089416504,
|
| 301 |
+
"learning_rate": 0.0002953958697394391,
|
| 302 |
+
"loss": 0.777,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.17565872020075282,
|
| 307 |
+
"grad_norm": 2.001302480697632,
|
| 308 |
+
"learning_rate": 0.000294841493050952,
|
| 309 |
+
"loss": 0.7797,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.17984107068172314,
|
| 314 |
+
"grad_norm": 0.7828574776649475,
|
| 315 |
+
"learning_rate": 0.0002942561914776124,
|
| 316 |
+
"loss": 0.7815,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.18402342116269343,
|
| 321 |
+
"grad_norm": 1.4854490756988525,
|
| 322 |
+
"learning_rate": 0.00029364008998626086,
|
| 323 |
+
"loss": 0.7608,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.18820577164366373,
|
| 328 |
+
"grad_norm": 1.1406800746917725,
|
| 329 |
+
"learning_rate": 0.00029299332011978107,
|
| 330 |
+
"loss": 0.747,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.19238812212463405,
|
| 335 |
+
"grad_norm": 1.7346460819244385,
|
| 336 |
+
"learning_rate": 0.00029231601996901433,
|
| 337 |
+
"loss": 0.7555,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.19657047260560434,
|
| 342 |
+
"grad_norm": 1.7754813432693481,
|
| 343 |
+
"learning_rate": 0.0002916083341432763,
|
| 344 |
+
"loss": 0.7626,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.20075282308657466,
|
| 349 |
+
"grad_norm": 1.2126661539077759,
|
| 350 |
+
"learning_rate": 0.00029087041373948135,
|
| 351 |
+
"loss": 0.7237,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.20493517356754495,
|
| 356 |
+
"grad_norm": 1.930538535118103,
|
| 357 |
+
"learning_rate": 0.00029010241630988217,
|
| 358 |
+
"loss": 0.7672,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.20911752404851527,
|
| 363 |
+
"grad_norm": 2.1792216300964355,
|
| 364 |
+
"learning_rate": 0.0002893045058284311,
|
| 365 |
+
"loss": 0.7416,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.21329987452948557,
|
| 370 |
+
"grad_norm": 1.416754961013794,
|
| 371 |
+
"learning_rate": 0.0002884768526557703,
|
| 372 |
+
"loss": 0.7196,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2174822250104559,
|
| 377 |
+
"grad_norm": 1.6103583574295044,
|
| 378 |
+
"learning_rate": 0.0002876196335028581,
|
| 379 |
+
"loss": 0.7397,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.22166457549142618,
|
| 384 |
+
"grad_norm": 1.0755459070205688,
|
| 385 |
+
"learning_rate": 0.0002867330313932402,
|
| 386 |
+
"loss": 0.7644,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.2258469259723965,
|
| 391 |
+
"grad_norm": 0.8303298354148865,
|
| 392 |
+
"learning_rate": 0.000285817235623972,
|
| 393 |
+
"loss": 0.7393,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2300292764533668,
|
| 398 |
+
"grad_norm": 1.4747998714447021,
|
| 399 |
+
"learning_rate": 0.00028487244172520246,
|
| 400 |
+
"loss": 0.7121,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.23421162693433709,
|
| 405 |
+
"grad_norm": 2.582953929901123,
|
| 406 |
+
"learning_rate": 0.0002838988514184267,
|
| 407 |
+
"loss": 0.7361,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.2383939774153074,
|
| 412 |
+
"grad_norm": 2.413325309753418,
|
| 413 |
+
"learning_rate": 0.0002828966725734167,
|
| 414 |
+
"loss": 0.74,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.2425763278962777,
|
| 419 |
+
"grad_norm": 0.7637856006622314,
|
| 420 |
+
"learning_rate": 0.0002818661191638393,
|
| 421 |
+
"loss": 0.7096,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.24675867837724802,
|
| 426 |
+
"grad_norm": 1.757056713104248,
|
| 427 |
+
"learning_rate": 0.0002808074112215711,
|
| 428 |
+
"loss": 0.7205,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.25094102885821834,
|
| 433 |
+
"grad_norm": 0.8766753077507019,
|
| 434 |
+
"learning_rate": 0.0002797207747897198,
|
| 435 |
+
"loss": 0.7098,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.2551233793391886,
|
| 440 |
+
"grad_norm": 1.449209213256836,
|
| 441 |
+
"learning_rate": 0.00027860644187436195,
|
| 442 |
+
"loss": 0.725,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.2593057298201589,
|
| 447 |
+
"grad_norm": 0.6825206875801086,
|
| 448 |
+
"learning_rate": 0.0002774646503950078,
|
| 449 |
+
"loss": 0.6938,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.26348808030112925,
|
| 454 |
+
"grad_norm": 1.119585394859314,
|
| 455 |
+
"learning_rate": 0.0002762956441338036,
|
| 456 |
+
"loss": 0.698,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.26767043078209957,
|
| 461 |
+
"grad_norm": 0.9425824880599976,
|
| 462 |
+
"learning_rate": 0.0002750996726834817,
|
| 463 |
+
"loss": 0.7189,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.27185278126306983,
|
| 468 |
+
"grad_norm": 0.5979897975921631,
|
| 469 |
+
"learning_rate": 0.0002738769913940706,
|
| 470 |
+
"loss": 0.7039,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.27603513174404015,
|
| 475 |
+
"grad_norm": 1.8769757747650146,
|
| 476 |
+
"learning_rate": 0.00027262786131837573,
|
| 477 |
+
"loss": 0.7035,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2802174822250105,
|
| 482 |
+
"grad_norm": 1.1395800113677979,
|
| 483 |
+
"learning_rate": 0.0002713525491562421,
|
| 484 |
+
"loss": 0.6898,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.28439983270598074,
|
| 489 |
+
"grad_norm": 1.0573526620864868,
|
| 490 |
+
"learning_rate": 0.0002700513271976119,
|
| 491 |
+
"loss": 0.7042,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.28858218318695106,
|
| 496 |
+
"grad_norm": 0.5185459852218628,
|
| 497 |
+
"learning_rate": 0.0002687244732643881,
|
| 498 |
+
"loss": 0.6914,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.2927645336679214,
|
| 503 |
+
"grad_norm": 2.7914602756500244,
|
| 504 |
+
"learning_rate": 0.0002673722706511174,
|
| 505 |
+
"loss": 0.7049,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2969468841488917,
|
| 510 |
+
"grad_norm": 3.0459792613983154,
|
| 511 |
+
"learning_rate": 0.000265995008064504,
|
| 512 |
+
"loss": 0.7148,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.30112923462986196,
|
| 517 |
+
"grad_norm": 2.1906723976135254,
|
| 518 |
+
"learning_rate": 0.00026459297956176885,
|
| 519 |
+
"loss": 0.7074,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3053115851108323,
|
| 524 |
+
"grad_norm": 1.6257227659225464,
|
| 525 |
+
"learning_rate": 0.00026316648448786536,
|
| 526 |
+
"loss": 0.6985,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3094939355918026,
|
| 531 |
+
"grad_norm": 0.7152910828590393,
|
| 532 |
+
"learning_rate": 0.00026171582741156725,
|
| 533 |
+
"loss": 0.6875,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.3136762860727729,
|
| 538 |
+
"grad_norm": 2.4449851512908936,
|
| 539 |
+
"learning_rate": 0.0002602413180604401,
|
| 540 |
+
"loss": 0.6787,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.3178586365537432,
|
| 545 |
+
"grad_norm": 0.5180588960647583,
|
| 546 |
+
"learning_rate": 0.000258743271254712,
|
| 547 |
+
"loss": 0.6724,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.3220409870347135,
|
| 552 |
+
"grad_norm": 1.5739381313323975,
|
| 553 |
+
"learning_rate": 0.00025722200684005715,
|
| 554 |
+
"loss": 0.7076,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.32622333751568383,
|
| 559 |
+
"grad_norm": 0.8701817989349365,
|
| 560 |
+
"learning_rate": 0.00025567784961930546,
|
| 561 |
+
"loss": 0.6841,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.3304056879966541,
|
| 566 |
+
"grad_norm": 1.474747896194458,
|
| 567 |
+
"learning_rate": 0.0002541111292830951,
|
| 568 |
+
"loss": 0.713,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.3345880384776244,
|
| 573 |
+
"grad_norm": 1.8884798288345337,
|
| 574 |
+
"learning_rate": 0.00025252218033947993,
|
| 575 |
+
"loss": 0.6893,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.33877038895859474,
|
| 580 |
+
"grad_norm": 0.8834472894668579,
|
| 581 |
+
"learning_rate": 0.00025091134204250997,
|
| 582 |
+
"loss": 0.6966,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.34295273943956506,
|
| 587 |
+
"grad_norm": 0.6324520707130432,
|
| 588 |
+
"learning_rate": 0.00024927895831979745,
|
| 589 |
+
"loss": 0.6882,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3471350899205353,
|
| 594 |
+
"grad_norm": 2.353163480758667,
|
| 595 |
+
"learning_rate": 0.00024762537769908535,
|
| 596 |
+
"loss": 0.6829,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.35131744040150564,
|
| 601 |
+
"grad_norm": 1.3682096004486084,
|
| 602 |
+
"learning_rate": 0.00024595095323383365,
|
| 603 |
+
"loss": 0.6912,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.35549979088247596,
|
| 608 |
+
"grad_norm": 0.9962055087089539,
|
| 609 |
+
"learning_rate": 0.0002442560424278399,
|
| 610 |
+
"loss": 0.6857,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.3596821413634463,
|
| 615 |
+
"grad_norm": 1.1282930374145508,
|
| 616 |
+
"learning_rate": 0.00024254100715890846,
|
| 617 |
+
"loss": 0.6696,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.36386449184441655,
|
| 622 |
+
"grad_norm": 0.934388279914856,
|
| 623 |
+
"learning_rate": 0.00024080621360158717,
|
| 624 |
+
"loss": 0.6841,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.36804684232538687,
|
| 629 |
+
"grad_norm": 1.4339077472686768,
|
| 630 |
+
"learning_rate": 0.00023905203214898558,
|
| 631 |
+
"loss": 0.6705,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.3722291928063572,
|
| 636 |
+
"grad_norm": 1.0309265851974487,
|
| 637 |
+
"learning_rate": 0.00023727883733369292,
|
| 638 |
+
"loss": 0.6706,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.37641154328732745,
|
| 643 |
+
"grad_norm": 1.9208811521530151,
|
| 644 |
+
"learning_rate": 0.00023548700774781242,
|
| 645 |
+
"loss": 0.6637,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3805938937682978,
|
| 650 |
+
"grad_norm": 1.0379974842071533,
|
| 651 |
+
"learning_rate": 0.00023367692596212858,
|
| 652 |
+
"loss": 0.68,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.3847762442492681,
|
| 657 |
+
"grad_norm": 1.852662444114685,
|
| 658 |
+
"learning_rate": 0.00023184897844442495,
|
| 659 |
+
"loss": 0.6589,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3889585947302384,
|
| 664 |
+
"grad_norm": 1.1750479936599731,
|
| 665 |
+
"learning_rate": 0.00023000355547697027,
|
| 666 |
+
"loss": 0.6675,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3931409452112087,
|
| 671 |
+
"grad_norm": 1.6473002433776855,
|
| 672 |
+
"learning_rate": 0.00022814105107318952,
|
| 673 |
+
"loss": 0.6709,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.397323295692179,
|
| 678 |
+
"grad_norm": 1.2356650829315186,
|
| 679 |
+
"learning_rate": 0.00022626186289353913,
|
| 680 |
+
"loss": 0.6652,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.4015056461731493,
|
| 685 |
+
"grad_norm": 1.1605840921401978,
|
| 686 |
+
"learning_rate": 0.00022436639216060275,
|
| 687 |
+
"loss": 0.6698,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.40568799665411964,
|
| 692 |
+
"grad_norm": 1.5935866832733154,
|
| 693 |
+
"learning_rate": 0.00022245504357342716,
|
| 694 |
+
"loss": 0.6688,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.4098703471350899,
|
| 699 |
+
"grad_norm": 0.810558557510376,
|
| 700 |
+
"learning_rate": 0.00022052822522111522,
|
| 701 |
+
"loss": 0.6524,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.41405269761606023,
|
| 706 |
+
"grad_norm": 0.7008018493652344,
|
| 707 |
+
"learning_rate": 0.00021858634849569576,
|
| 708 |
+
"loss": 0.6924,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.41823504809703055,
|
| 713 |
+
"grad_norm": 1.7558863162994385,
|
| 714 |
+
"learning_rate": 0.0002166298280042877,
|
| 715 |
+
"loss": 0.6711,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.4224173985780008,
|
| 720 |
+
"grad_norm": 1.573688268661499,
|
| 721 |
+
"learning_rate": 0.00021465908148057787,
|
| 722 |
+
"loss": 0.6674,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.42659974905897113,
|
| 727 |
+
"grad_norm": 1.4761265516281128,
|
| 728 |
+
"learning_rate": 0.00021267452969563153,
|
| 729 |
+
"loss": 0.6706,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.43078209953994145,
|
| 734 |
+
"grad_norm": 1.7749208211898804,
|
| 735 |
+
"learning_rate": 0.00021067659636805403,
|
| 736 |
+
"loss": 0.6469,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.4349644500209118,
|
| 741 |
+
"grad_norm": 1.0164939165115356,
|
| 742 |
+
"learning_rate": 0.00020866570807352337,
|
| 743 |
+
"loss": 0.6764,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.43914680050188204,
|
| 748 |
+
"grad_norm": 1.6237319707870483,
|
| 749 |
+
"learning_rate": 0.00020664229415371266,
|
| 750 |
+
"loss": 0.6694,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.44332915098285236,
|
| 755 |
+
"grad_norm": 1.5586035251617432,
|
| 756 |
+
"learning_rate": 0.00020460678662462194,
|
| 757 |
+
"loss": 0.6562,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.4475115014638227,
|
| 762 |
+
"grad_norm": 1.771645188331604,
|
| 763 |
+
"learning_rate": 0.0002025596200843394,
|
| 764 |
+
"loss": 0.6622,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.451693851944793,
|
| 769 |
+
"grad_norm": 0.5951160788536072,
|
| 770 |
+
"learning_rate": 0.0002005012316202506,
|
| 771 |
+
"loss": 0.651,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.45587620242576327,
|
| 776 |
+
"grad_norm": 0.793093740940094,
|
| 777 |
+
"learning_rate": 0.00019843206071571692,
|
| 778 |
+
"loss": 0.6634,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4600585529067336,
|
| 783 |
+
"grad_norm": 1.0352814197540283,
|
| 784 |
+
"learning_rate": 0.0001963525491562421,
|
| 785 |
+
"loss": 0.6636,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.4642409033877039,
|
| 790 |
+
"grad_norm": 0.843008816242218,
|
| 791 |
+
"learning_rate": 0.00019426314093514717,
|
| 792 |
+
"loss": 0.6407,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.46842325386867417,
|
| 797 |
+
"grad_norm": 1.7540709972381592,
|
| 798 |
+
"learning_rate": 0.00019216428215877425,
|
| 799 |
+
"loss": 0.638,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.4726056043496445,
|
| 804 |
+
"grad_norm": 0.5828922390937805,
|
| 805 |
+
"learning_rate": 0.00019005642095123895,
|
| 806 |
+
"loss": 0.6625,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.4767879548306148,
|
| 811 |
+
"grad_norm": 0.7700462937355042,
|
| 812 |
+
"learning_rate": 0.00018794000735875208,
|
| 813 |
+
"loss": 0.6428,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.48097030531158513,
|
| 818 |
+
"grad_norm": 0.8344655632972717,
|
| 819 |
+
"learning_rate": 0.00018581549325353126,
|
| 820 |
+
"loss": 0.6553,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.4851526557925554,
|
| 825 |
+
"grad_norm": 1.2676873207092285,
|
| 826 |
+
"learning_rate": 0.000183683332237322,
|
| 827 |
+
"loss": 0.6645,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.4893350062735257,
|
| 832 |
+
"grad_norm": 1.4888837337493896,
|
| 833 |
+
"learning_rate": 0.00018154397954454993,
|
| 834 |
+
"loss": 0.6859,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.49351735675449604,
|
| 839 |
+
"grad_norm": 0.7020601034164429,
|
| 840 |
+
"learning_rate": 0.00017939789194512472,
|
| 841 |
+
"loss": 0.6456,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.49769970723546636,
|
| 846 |
+
"grad_norm": 1.1964813470840454,
|
| 847 |
+
"learning_rate": 0.00017724552764691545,
|
| 848 |
+
"loss": 0.6594,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.5018820577164367,
|
| 853 |
+
"grad_norm": 1.1332772970199585,
|
| 854 |
+
"learning_rate": 0.00017508734619791966,
|
| 855 |
+
"loss": 0.6606,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.506064408197407,
|
| 860 |
+
"grad_norm": 1.6122368574142456,
|
| 861 |
+
"learning_rate": 0.00017292380838814577,
|
| 862 |
+
"loss": 0.6468,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.5102467586783772,
|
| 867 |
+
"grad_norm": 0.8950415849685669,
|
| 868 |
+
"learning_rate": 0.00017075537615123042,
|
| 869 |
+
"loss": 0.6615,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.5144291091593476,
|
| 874 |
+
"grad_norm": 1.9138753414154053,
|
| 875 |
+
"learning_rate": 0.00016858251246581216,
|
| 876 |
+
"loss": 0.6683,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.5186114596403179,
|
| 881 |
+
"grad_norm": 0.9320158362388611,
|
| 882 |
+
"learning_rate": 0.00016640568125668117,
|
| 883 |
+
"loss": 0.6734,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.5227938101212881,
|
| 888 |
+
"grad_norm": 1.2331713438034058,
|
| 889 |
+
"learning_rate": 0.00016422534729572738,
|
| 890 |
+
"loss": 0.6582,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.5269761606022585,
|
| 895 |
+
"grad_norm": 1.1182340383529663,
|
| 896 |
+
"learning_rate": 0.00016204197610270816,
|
| 897 |
+
"loss": 0.6533,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.5311585110832288,
|
| 902 |
+
"grad_norm": 0.6500148773193359,
|
| 903 |
+
"learning_rate": 0.00015985603384585542,
|
| 904 |
+
"loss": 0.6396,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.5353408615641991,
|
| 909 |
+
"grad_norm": 0.9531376361846924,
|
| 910 |
+
"learning_rate": 0.00015766798724234506,
|
| 911 |
+
"loss": 0.6337,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.5395232120451694,
|
| 916 |
+
"grad_norm": 0.7729771733283997,
|
| 917 |
+
"learning_rate": 0.00015547830345864885,
|
| 918 |
+
"loss": 0.6498,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.5437055625261397,
|
| 923 |
+
"grad_norm": 0.6831007599830627,
|
| 924 |
+
"learning_rate": 0.0001532874500107902,
|
| 925 |
+
"loss": 0.6404,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.54788791300711,
|
| 930 |
+
"grad_norm": 1.2160038948059082,
|
| 931 |
+
"learning_rate": 0.00015109589466452594,
|
| 932 |
+
"loss": 0.658,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.5520702634880803,
|
| 937 |
+
"grad_norm": 1.015773057937622,
|
| 938 |
+
"learning_rate": 0.00014890410533547404,
|
| 939 |
+
"loss": 0.6507,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.5562526139690506,
|
| 944 |
+
"grad_norm": 1.208256721496582,
|
| 945 |
+
"learning_rate": 0.00014671254998920976,
|
| 946 |
+
"loss": 0.6399,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.560434964450021,
|
| 951 |
+
"grad_norm": 0.7431871294975281,
|
| 952 |
+
"learning_rate": 0.00014452169654135115,
|
| 953 |
+
"loss": 0.6534,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.5646173149309912,
|
| 958 |
+
"grad_norm": 0.6661595702171326,
|
| 959 |
+
"learning_rate": 0.00014233201275765494,
|
| 960 |
+
"loss": 0.6343,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5687996654119615,
|
| 965 |
+
"grad_norm": 1.2753201723098755,
|
| 966 |
+
"learning_rate": 0.00014014396615414458,
|
| 967 |
+
"loss": 0.6296,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5729820158929319,
|
| 972 |
+
"grad_norm": 1.4110949039459229,
|
| 973 |
+
"learning_rate": 0.00013795802389729184,
|
| 974 |
+
"loss": 0.6452,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.5771643663739021,
|
| 979 |
+
"grad_norm": 1.4824358224868774,
|
| 980 |
+
"learning_rate": 0.00013577465270427262,
|
| 981 |
+
"loss": 0.6348,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.5813467168548725,
|
| 986 |
+
"grad_norm": 1.8900264501571655,
|
| 987 |
+
"learning_rate": 0.00013359431874331886,
|
| 988 |
+
"loss": 0.6509,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.5855290673358428,
|
| 993 |
+
"grad_norm": 1.652632236480713,
|
| 994 |
+
"learning_rate": 0.0001314174875341878,
|
| 995 |
+
"loss": 0.6206,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.589711417816813,
|
| 1000 |
+
"grad_norm": 1.1248772144317627,
|
| 1001 |
+
"learning_rate": 0.00012924462384876953,
|
| 1002 |
+
"loss": 0.6299,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.5938937682977834,
|
| 1007 |
+
"grad_norm": 0.7448098659515381,
|
| 1008 |
+
"learning_rate": 0.00012707619161185423,
|
| 1009 |
+
"loss": 0.6483,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5980761187787537,
|
| 1014 |
+
"grad_norm": 0.6708864569664001,
|
| 1015 |
+
"learning_rate": 0.00012491265380208032,
|
| 1016 |
+
"loss": 0.6473,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.6022584692597239,
|
| 1021 |
+
"grad_norm": 0.8381022810935974,
|
| 1022 |
+
"learning_rate": 0.00012275447235308453,
|
| 1023 |
+
"loss": 0.6356,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.6064408197406943,
|
| 1028 |
+
"grad_norm": 1.3462333679199219,
|
| 1029 |
+
"learning_rate": 0.00012060210805487529,
|
| 1030 |
+
"loss": 0.6388,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.6106231702216646,
|
| 1035 |
+
"grad_norm": 1.2015129327774048,
|
| 1036 |
+
"learning_rate": 0.00011845602045545008,
|
| 1037 |
+
"loss": 0.6258,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.6148055207026348,
|
| 1042 |
+
"grad_norm": 0.7825962901115417,
|
| 1043 |
+
"learning_rate": 0.00011631666776267803,
|
| 1044 |
+
"loss": 0.6401,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.6189878711836052,
|
| 1049 |
+
"grad_norm": 0.9470372200012207,
|
| 1050 |
+
"learning_rate": 0.00011418450674646868,
|
| 1051 |
+
"loss": 0.6501,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.6231702216645755,
|
| 1056 |
+
"grad_norm": 0.9243600368499756,
|
| 1057 |
+
"learning_rate": 0.00011205999264124786,
|
| 1058 |
+
"loss": 0.6195,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.6273525721455459,
|
| 1063 |
+
"grad_norm": 1.931402325630188,
|
| 1064 |
+
"learning_rate": 0.00010994357904876106,
|
| 1065 |
+
"loss": 0.6264,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.6315349226265161,
|
| 1070 |
+
"grad_norm": 1.7754958868026733,
|
| 1071 |
+
"learning_rate": 0.00010783571784122577,
|
| 1072 |
+
"loss": 0.6351,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.6357172731074864,
|
| 1077 |
+
"grad_norm": 0.627479076385498,
|
| 1078 |
+
"learning_rate": 0.00010573685906485282,
|
| 1079 |
+
"loss": 0.6395,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.6398996235884568,
|
| 1084 |
+
"grad_norm": 2.4568421840667725,
|
| 1085 |
+
"learning_rate": 0.0001036474508437579,
|
| 1086 |
+
"loss": 0.6257,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.644081974069427,
|
| 1091 |
+
"grad_norm": 1.9016671180725098,
|
| 1092 |
+
"learning_rate": 0.0001015679392842831,
|
| 1093 |
+
"loss": 0.6446,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.6482643245503973,
|
| 1098 |
+
"grad_norm": 0.7173120975494385,
|
| 1099 |
+
"learning_rate": 9.949876837974944e-05,
|
| 1100 |
+
"loss": 0.6312,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.6524466750313677,
|
| 1105 |
+
"grad_norm": 1.9048292636871338,
|
| 1106 |
+
"learning_rate": 9.744037991566058e-05,
|
| 1107 |
+
"loss": 0.622,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.6566290255123379,
|
| 1112 |
+
"grad_norm": 1.2454332113265991,
|
| 1113 |
+
"learning_rate": 9.5393213375378e-05,
|
| 1114 |
+
"loss": 0.6219,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.6608113759933082,
|
| 1119 |
+
"grad_norm": 1.0283215045928955,
|
| 1120 |
+
"learning_rate": 9.33577058462873e-05,
|
| 1121 |
+
"loss": 0.6236,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.6649937264742786,
|
| 1126 |
+
"grad_norm": 1.8116226196289062,
|
| 1127 |
+
"learning_rate": 9.133429192647661e-05,
|
| 1128 |
+
"loss": 0.6244,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.6691760769552488,
|
| 1133 |
+
"grad_norm": 0.8519335389137268,
|
| 1134 |
+
"learning_rate": 8.932340363194595e-05,
|
| 1135 |
+
"loss": 0.6253,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.6733584274362192,
|
| 1140 |
+
"grad_norm": 1.004647135734558,
|
| 1141 |
+
"learning_rate": 8.73254703043685e-05,
|
| 1142 |
+
"loss": 0.6278,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.6775407779171895,
|
| 1147 |
+
"grad_norm": 0.945331871509552,
|
| 1148 |
+
"learning_rate": 8.534091851942214e-05,
|
| 1149 |
+
"loss": 0.6251,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.6817231283981597,
|
| 1154 |
+
"grad_norm": 0.4643709659576416,
|
| 1155 |
+
"learning_rate": 8.337017199571235e-05,
|
| 1156 |
+
"loss": 0.6298,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6859054788791301,
|
| 1161 |
+
"grad_norm": 0.8933060169219971,
|
| 1162 |
+
"learning_rate": 8.141365150430421e-05,
|
| 1163 |
+
"loss": 0.6419,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6900878293601004,
|
| 1168 |
+
"grad_norm": 2.5518875122070312,
|
| 1169 |
+
"learning_rate": 7.947177477888472e-05,
|
| 1170 |
+
"loss": 0.6424,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.6942701798410706,
|
| 1175 |
+
"grad_norm": 0.7830976247787476,
|
| 1176 |
+
"learning_rate": 7.754495642657282e-05,
|
| 1177 |
+
"loss": 0.6292,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.698452530322041,
|
| 1182 |
+
"grad_norm": 1.2565546035766602,
|
| 1183 |
+
"learning_rate": 7.563360783939722e-05,
|
| 1184 |
+
"loss": 0.6308,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.7026348808030113,
|
| 1189 |
+
"grad_norm": 1.588156819343567,
|
| 1190 |
+
"learning_rate": 7.373813710646083e-05,
|
| 1191 |
+
"loss": 0.6249,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.7068172312839816,
|
| 1196 |
+
"grad_norm": 0.6486766934394836,
|
| 1197 |
+
"learning_rate": 7.185894892681048e-05,
|
| 1198 |
+
"loss": 0.6308,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.7109995817649519,
|
| 1203 |
+
"grad_norm": 0.7454473972320557,
|
| 1204 |
+
"learning_rate": 6.999644452302975e-05,
|
| 1205 |
+
"loss": 0.6267,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.7151819322459222,
|
| 1210 |
+
"grad_norm": 0.6658061146736145,
|
| 1211 |
+
"learning_rate": 6.815102155557501e-05,
|
| 1212 |
+
"loss": 0.6162,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.7193642827268926,
|
| 1217 |
+
"grad_norm": 1.0052908658981323,
|
| 1218 |
+
"learning_rate": 6.632307403787138e-05,
|
| 1219 |
+
"loss": 0.644,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.7235466332078628,
|
| 1224 |
+
"grad_norm": 0.7472626566886902,
|
| 1225 |
+
"learning_rate": 6.451299225218754e-05,
|
| 1226 |
+
"loss": 0.616,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.7277289836888331,
|
| 1231 |
+
"grad_norm": 0.587893009185791,
|
| 1232 |
+
"learning_rate": 6.27211626663071e-05,
|
| 1233 |
+
"loss": 0.6318,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.7319113341698035,
|
| 1238 |
+
"grad_norm": 0.898607611656189,
|
| 1239 |
+
"learning_rate": 6.0947967851014405e-05,
|
| 1240 |
+
"loss": 0.6409,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.7360936846507737,
|
| 1245 |
+
"grad_norm": 0.7444003224372864,
|
| 1246 |
+
"learning_rate": 5.919378639841281e-05,
|
| 1247 |
+
"loss": 0.6214,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.740276035131744,
|
| 1252 |
+
"grad_norm": 1.199029564857483,
|
| 1253 |
+
"learning_rate": 5.745899284109154e-05,
|
| 1254 |
+
"loss": 0.6184,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.7444583856127144,
|
| 1259 |
+
"grad_norm": 1.2874826192855835,
|
| 1260 |
+
"learning_rate": 5.57439575721601e-05,
|
| 1261 |
+
"loss": 0.6233,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.7486407360936846,
|
| 1266 |
+
"grad_norm": 1.3848364353179932,
|
| 1267 |
+
"learning_rate": 5.4049046766166335e-05,
|
| 1268 |
+
"loss": 0.6043,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.7528230865746549,
|
| 1273 |
+
"grad_norm": 0.6463631987571716,
|
| 1274 |
+
"learning_rate": 5.237462230091467e-05,
|
| 1275 |
+
"loss": 0.6361,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.7570054370556253,
|
| 1280 |
+
"grad_norm": 1.0429089069366455,
|
| 1281 |
+
"learning_rate": 5.07210416802025e-05,
|
| 1282 |
+
"loss": 0.6206,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.7611877875365956,
|
| 1287 |
+
"grad_norm": 0.7253705263137817,
|
| 1288 |
+
"learning_rate": 4.908865795748999e-05,
|
| 1289 |
+
"loss": 0.6312,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.7653701380175659,
|
| 1294 |
+
"grad_norm": 0.498542457818985,
|
| 1295 |
+
"learning_rate": 4.74778196605201e-05,
|
| 1296 |
+
"loss": 0.6421,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.7695524884985362,
|
| 1301 |
+
"grad_norm": 0.6464426517486572,
|
| 1302 |
+
"learning_rate": 4.58888707169049e-05,
|
| 1303 |
+
"loss": 0.6047,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.7737348389795065,
|
| 1308 |
+
"grad_norm": 0.6658442616462708,
|
| 1309 |
+
"learning_rate": 4.432215038069449e-05,
|
| 1310 |
+
"loss": 0.623,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.7779171894604768,
|
| 1315 |
+
"grad_norm": 0.6936389207839966,
|
| 1316 |
+
"learning_rate": 4.277799315994286e-05,
|
| 1317 |
+
"loss": 0.6226,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.7820995399414471,
|
| 1322 |
+
"grad_norm": 0.5900949835777283,
|
| 1323 |
+
"learning_rate": 4.125672874528797e-05,
|
| 1324 |
+
"loss": 0.6314,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.7862818904224174,
|
| 1329 |
+
"grad_norm": 0.9340611100196838,
|
| 1330 |
+
"learning_rate": 3.97586819395599e-05,
|
| 1331 |
+
"loss": 0.6252,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.7904642409033877,
|
| 1336 |
+
"grad_norm": 1.272733211517334,
|
| 1337 |
+
"learning_rate": 3.8284172588432716e-05,
|
| 1338 |
+
"loss": 0.6236,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.794646591384358,
|
| 1343 |
+
"grad_norm": 0.7782835364341736,
|
| 1344 |
+
"learning_rate": 3.6833515512134606e-05,
|
| 1345 |
+
"loss": 0.6096,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7988289418653283,
|
| 1350 |
+
"grad_norm": 0.6020464301109314,
|
| 1351 |
+
"learning_rate": 3.540702043823113e-05,
|
| 1352 |
+
"loss": 0.6124,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.8030112923462986,
|
| 1357 |
+
"grad_norm": 0.675359845161438,
|
| 1358 |
+
"learning_rate": 3.4004991935496004e-05,
|
| 1359 |
+
"loss": 0.5955,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.8071936428272689,
|
| 1364 |
+
"grad_norm": 0.5639395117759705,
|
| 1365 |
+
"learning_rate": 3.262772934888265e-05,
|
| 1366 |
+
"loss": 0.6069,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.8113759933082393,
|
| 1371 |
+
"grad_norm": 0.4711320698261261,
|
| 1372 |
+
"learning_rate": 3.1275526735611896e-05,
|
| 1373 |
+
"loss": 0.6102,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.8155583437892095,
|
| 1378 |
+
"grad_norm": 1.0711911916732788,
|
| 1379 |
+
"learning_rate": 2.9948672802388135e-05,
|
| 1380 |
+
"loss": 0.6391,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.8197406942701798,
|
| 1385 |
+
"grad_norm": 0.4709874987602234,
|
| 1386 |
+
"learning_rate": 2.8647450843757897e-05,
|
| 1387 |
+
"loss": 0.6186,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.8239230447511502,
|
| 1392 |
+
"grad_norm": 0.5402533411979675,
|
| 1393 |
+
"learning_rate": 2.7372138681624244e-05,
|
| 1394 |
+
"loss": 0.613,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.8281053952321205,
|
| 1399 |
+
"grad_norm": 0.6864106059074402,
|
| 1400 |
+
"learning_rate": 2.6123008605929375e-05,
|
| 1401 |
+
"loss": 0.6215,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.8322877457130907,
|
| 1406 |
+
"grad_norm": 0.5939123630523682,
|
| 1407 |
+
"learning_rate": 2.4900327316518326e-05,
|
| 1408 |
+
"loss": 0.6168,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.8364700961940611,
|
| 1413 |
+
"grad_norm": 0.7122395038604736,
|
| 1414 |
+
"learning_rate": 2.3704355866196373e-05,
|
| 1415 |
+
"loss": 0.6053,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.8406524466750314,
|
| 1420 |
+
"grad_norm": 0.6920621395111084,
|
| 1421 |
+
"learning_rate": 2.2535349604992153e-05,
|
| 1422 |
+
"loss": 0.6097,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.8448347971560016,
|
| 1427 |
+
"grad_norm": 1.7864691019058228,
|
| 1428 |
+
"learning_rate": 2.1393558125638066e-05,
|
| 1429 |
+
"loss": 0.6382,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.849017147636972,
|
| 1434 |
+
"grad_norm": 0.7472600936889648,
|
| 1435 |
+
"learning_rate": 2.027922521028018e-05,
|
| 1436 |
+
"loss": 0.6159,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.8531994981179423,
|
| 1441 |
+
"grad_norm": 1.722579836845398,
|
| 1442 |
+
"learning_rate": 1.9192588778428842e-05,
|
| 1443 |
+
"loss": 0.6011,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.8573818485989126,
|
| 1448 |
+
"grad_norm": 0.5121240019798279,
|
| 1449 |
+
"learning_rate": 1.813388083616068e-05,
|
| 1450 |
+
"loss": 0.6031,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.8615641990798829,
|
| 1455 |
+
"grad_norm": 0.5083288550376892,
|
| 1456 |
+
"learning_rate": 1.7103327426583265e-05,
|
| 1457 |
+
"loss": 0.5845,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.8657465495608532,
|
| 1462 |
+
"grad_norm": 0.46453267335891724,
|
| 1463 |
+
"learning_rate": 1.6101148581573274e-05,
|
| 1464 |
+
"loss": 0.6031,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.8699289000418235,
|
| 1469 |
+
"grad_norm": 0.8352246880531311,
|
| 1470 |
+
"learning_rate": 1.5127558274797535e-05,
|
| 1471 |
+
"loss": 0.6024,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.8741112505227938,
|
| 1476 |
+
"grad_norm": 0.8040021061897278,
|
| 1477 |
+
"learning_rate": 1.4182764376028006e-05,
|
| 1478 |
+
"loss": 0.635,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.8782936010037641,
|
| 1483 |
+
"grad_norm": 0.7152721881866455,
|
| 1484 |
+
"learning_rate": 1.326696860675981e-05,
|
| 1485 |
+
"loss": 0.6162,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.8824759514847345,
|
| 1490 |
+
"grad_norm": 0.6833348274230957,
|
| 1491 |
+
"learning_rate": 1.2380366497141886e-05,
|
| 1492 |
+
"loss": 0.6217,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.8866583019657047,
|
| 1497 |
+
"grad_norm": 0.6803585886955261,
|
| 1498 |
+
"learning_rate": 1.1523147344229716e-05,
|
| 1499 |
+
"loss": 0.6218,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.890840652446675,
|
| 1504 |
+
"grad_norm": 0.9396490454673767,
|
| 1505 |
+
"learning_rate": 1.069549417156887e-05,
|
| 1506 |
+
"loss": 0.6176,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.8950230029276454,
|
| 1511 |
+
"grad_norm": 0.7006051540374756,
|
| 1512 |
+
"learning_rate": 9.89758369011781e-06,
|
| 1513 |
+
"loss": 0.6123,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.8992053534086156,
|
| 1518 |
+
"grad_norm": 0.5886880159378052,
|
| 1519 |
+
"learning_rate": 9.129586260518634e-06,
|
| 1520 |
+
"loss": 0.5923,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.903387703889586,
|
| 1525 |
+
"grad_norm": 0.5032349228858948,
|
| 1526 |
+
"learning_rate": 8.391665856723655e-06,
|
| 1527 |
+
"loss": 0.619,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.9075700543705563,
|
| 1532 |
+
"grad_norm": 0.51763916015625,
|
| 1533 |
+
"learning_rate": 7.683980030985654e-06,
|
| 1534 |
+
"loss": 0.6039,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.9117524048515265,
|
| 1539 |
+
"grad_norm": 0.588211715221405,
|
| 1540 |
+
"learning_rate": 7.006679880218974e-06,
|
| 1541 |
+
"loss": 0.6057,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.9159347553324969,
|
| 1546 |
+
"grad_norm": 0.6631506681442261,
|
| 1547 |
+
"learning_rate": 6.359910013739122e-06,
|
| 1548 |
+
"loss": 0.6106,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.9201171058134672,
|
| 1553 |
+
"grad_norm": 0.5523665547370911,
|
| 1554 |
+
"learning_rate": 5.743808522387544e-06,
|
| 1555 |
+
"loss": 0.6058,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.9242994562944374,
|
| 1560 |
+
"grad_norm": 0.5903011560440063,
|
| 1561 |
+
"learning_rate": 5.158506949047975e-06,
|
| 1562 |
+
"loss": 0.6321,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.9284818067754078,
|
| 1567 |
+
"grad_norm": 0.6151677966117859,
|
| 1568 |
+
"learning_rate": 4.604130260560873e-06,
|
| 1569 |
+
"loss": 0.6171,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.9326641572563781,
|
| 1574 |
+
"grad_norm": 0.5324861407279968,
|
| 1575 |
+
"learning_rate": 4.080796821042082e-06,
|
| 1576 |
+
"loss": 0.6184,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.9368465077373483,
|
| 1581 |
+
"grad_norm": 0.5442612171173096,
|
| 1582 |
+
"learning_rate": 3.5886183666109405e-06,
|
| 1583 |
+
"loss": 0.6069,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.9410288582183187,
|
| 1588 |
+
"grad_norm": 0.4981847405433655,
|
| 1589 |
+
"learning_rate": 3.1276999815337544e-06,
|
| 1590 |
+
"loss": 0.62,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.945211208699289,
|
| 1595 |
+
"grad_norm": 0.45795848965644836,
|
| 1596 |
+
"learning_rate": 2.6981400757874584e-06,
|
| 1597 |
+
"loss": 0.6027,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.9493935591802594,
|
| 1602 |
+
"grad_norm": 0.78538978099823,
|
| 1603 |
+
"learning_rate": 2.3000303640481386e-06,
|
| 1604 |
+
"loss": 0.6084,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.9535759096612296,
|
| 1609 |
+
"grad_norm": 0.5453292727470398,
|
| 1610 |
+
"learning_rate": 1.9334558461092663e-06,
|
| 1611 |
+
"loss": 0.6043,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.9577582601421999,
|
| 1616 |
+
"grad_norm": 0.5924518704414368,
|
| 1617 |
+
"learning_rate": 1.598494788733462e-06,
|
| 1618 |
+
"loss": 0.6033,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.9619406106231703,
|
| 1623 |
+
"grad_norm": 0.5484139323234558,
|
| 1624 |
+
"learning_rate": 1.2952187089419642e-06,
|
| 1625 |
+
"loss": 0.616,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.9661229611041405,
|
| 1630 |
+
"grad_norm": 0.5363529324531555,
|
| 1631 |
+
"learning_rate": 1.0236923587450263e-06,
|
| 1632 |
+
"loss": 0.6196,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.9703053115851108,
|
| 1637 |
+
"grad_norm": 0.6777392625808716,
|
| 1638 |
+
"learning_rate": 7.839737113168931e-07,
|
| 1639 |
+
"loss": 0.6102,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.9744876620660812,
|
| 1644 |
+
"grad_norm": 0.5920884609222412,
|
| 1645 |
+
"learning_rate": 5.761139486180178e-07,
|
| 1646 |
+
"loss": 0.6075,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.9786700125470514,
|
| 1651 |
+
"grad_norm": 0.4196658730506897,
|
| 1652 |
+
"learning_rate": 4.0015745046725336e-07,
|
| 1653 |
+
"loss": 0.6017,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.9828523630280217,
|
| 1658 |
+
"grad_norm": 0.6661626100540161,
|
| 1659 |
+
"learning_rate": 2.5614178506644934e-07,
|
| 1660 |
+
"loss": 0.5787,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.9870347135089921,
|
| 1665 |
+
"grad_norm": 0.6183582544326782,
|
| 1666 |
+
"learning_rate": 1.4409770097926765e-07,
|
| 1667 |
+
"loss": 0.6188,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.9912170639899623,
|
| 1672 |
+
"grad_norm": 0.5524072051048279,
|
| 1673 |
+
"learning_rate": 6.40491205661009e-08,
|
| 1674 |
+
"loss": 0.6089,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.9953994144709327,
|
| 1679 |
+
"grad_norm": 0.5690078735351562,
|
| 1680 |
+
"learning_rate": 1.6013134876491362e-08,
|
| 1681 |
+
"loss": 0.6079,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.999581764951903,
|
| 1686 |
+
"grad_norm": 0.4828049838542938,
|
| 1687 |
+
"learning_rate": 0.0,
|
| 1688 |
+
"loss": 0.5901,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.999581764951903,
|
| 1693 |
+
"eval_loss": 1.2799842357635498,
|
| 1694 |
+
"eval_runtime": 0.8401,
|
| 1695 |
+
"eval_samples_per_second": 5.951,
|
| 1696 |
+
"eval_steps_per_second": 1.19,
|
| 1697 |
+
"step": 1195
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 0.999581764951903,
|
| 1701 |
+
"step": 1195,
|
| 1702 |
+
"total_flos": 9.109418934146171e+17,
|
| 1703 |
+
"train_loss": 1.440422640086218,
|
| 1704 |
+
"train_runtime": 6570.4384,
|
| 1705 |
+
"train_samples_per_second": 2.911,
|
| 1706 |
+
"train_steps_per_second": 0.182
|
| 1707 |
+
}
|
| 1708 |
+
],
|
| 1709 |
+
"logging_steps": 5,
|
| 1710 |
+
"max_steps": 1195,
|
| 1711 |
+
"num_input_tokens_seen": 0,
|
| 1712 |
+
"num_train_epochs": 1,
|
| 1713 |
+
"save_steps": 100,
|
| 1714 |
+
"stateful_callbacks": {
|
| 1715 |
+
"TrainerControl": {
|
| 1716 |
+
"args": {
|
| 1717 |
+
"should_epoch_stop": false,
|
| 1718 |
+
"should_evaluate": false,
|
| 1719 |
+
"should_log": false,
|
| 1720 |
+
"should_save": true,
|
| 1721 |
+
"should_training_stop": true
|
| 1722 |
+
},
|
| 1723 |
+
"attributes": {}
|
| 1724 |
+
}
|
| 1725 |
+
},
|
| 1726 |
+
"total_flos": 9.109418934146171e+17,
|
| 1727 |
+
"train_batch_size": 1,
|
| 1728 |
+
"trial_name": null,
|
| 1729 |
+
"trial_params": null
|
| 1730 |
+
}
|