Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gemma
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: google/gemma-7b
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: gemma7b-summarize-gpt4o-16k
|
| 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-summarize-gpt4o-16k
|
| 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: 4.9065
|
| 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: 0.0002
|
| 43 |
+
- train_batch_size: 4
|
| 44 |
+
- eval_batch_size: 4
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 2
|
| 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: 10
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 1.2436 | 1.0 | 102 | 2.5897 |
|
| 61 |
+
| 0.9498 | 2.0 | 204 | 2.4372 |
|
| 62 |
+
| 0.8227 | 3.0 | 306 | 2.4782 |
|
| 63 |
+
| 0.7124 | 4.0 | 408 | 2.6424 |
|
| 64 |
+
| 0.5905 | 5.0 | 510 | 3.0014 |
|
| 65 |
+
| 0.4741 | 6.0 | 612 | 3.4287 |
|
| 66 |
+
| 0.367 | 7.0 | 714 | 3.9945 |
|
| 67 |
+
| 0.3034 | 8.0 | 816 | 4.5447 |
|
| 68 |
+
| 0.265 | 9.0 | 918 | 4.8719 |
|
| 69 |
+
| 0.2597 | 10.0 | 1020 | 4.9065 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- PEFT 0.10.0
|
| 75 |
+
- Transformers 4.40.0
|
| 76 |
+
- Pytorch 2.2.1+cu121
|
| 77 |
+
- Datasets 2.18.0
|
| 78 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 50056096
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b6258118bda483e1a5424bd0cfecdb2a98bd9b68d32365ee5b0a968c565626e
|
| 3 |
size 50056096
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 7.797253181045473e+17,
|
| 4 |
+
"train_loss": 1.7479132745780197,
|
| 5 |
+
"train_runtime": 6650.2001,
|
| 6 |
+
"train_samples": 14998,
|
| 7 |
+
"train_samples_per_second": 2.45,
|
| 8 |
+
"train_steps_per_second": 0.153
|
| 9 |
+
}
|
runs/Jun08_15-10-36_user-WS-C621E-SAGE-Series/events.out.tfevents.1717827095.user-WS-C621E-SAGE-Series.12245.0
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 sha256:077f41166be99a32f70fde9cc65c686a784faaaf5bb7ce2ad8d5b5eefdaa9740
|
| 3 |
+
size 51642
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 7.797253181045473e+17,
|
| 4 |
+
"train_loss": 1.7479132745780197,
|
| 5 |
+
"train_runtime": 6650.2001,
|
| 6 |
+
"train_samples": 14998,
|
| 7 |
+
"train_samples_per_second": 2.45,
|
| 8 |
+
"train_steps_per_second": 0.153
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1545 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1020,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.00980392156862745,
|
| 13 |
+
"grad_norm": 496.0,
|
| 14 |
+
"learning_rate": 1.96078431372549e-06,
|
| 15 |
+
"loss": 48.0789,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.049019607843137254,
|
| 20 |
+
"grad_norm": 412.0,
|
| 21 |
+
"learning_rate": 9.803921568627451e-06,
|
| 22 |
+
"loss": 45.9094,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09803921568627451,
|
| 27 |
+
"grad_norm": 362.0,
|
| 28 |
+
"learning_rate": 1.9607843137254903e-05,
|
| 29 |
+
"loss": 43.1303,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.14705882352941177,
|
| 34 |
+
"grad_norm": 84.5,
|
| 35 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 36 |
+
"loss": 29.1951,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.19607843137254902,
|
| 41 |
+
"grad_norm": 46.0,
|
| 42 |
+
"learning_rate": 3.9215686274509805e-05,
|
| 43 |
+
"loss": 24.2648,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.24509803921568626,
|
| 48 |
+
"grad_norm": 17.75,
|
| 49 |
+
"learning_rate": 4.901960784313725e-05,
|
| 50 |
+
"loss": 21.3892,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.29411764705882354,
|
| 55 |
+
"grad_norm": 7.0625,
|
| 56 |
+
"learning_rate": 5.882352941176471e-05,
|
| 57 |
+
"loss": 19.4705,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.3431372549019608,
|
| 62 |
+
"grad_norm": 9.5,
|
| 63 |
+
"learning_rate": 6.862745098039216e-05,
|
| 64 |
+
"loss": 18.7197,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.39215686274509803,
|
| 69 |
+
"grad_norm": 19.5,
|
| 70 |
+
"learning_rate": 7.843137254901961e-05,
|
| 71 |
+
"loss": 17.5814,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.4411764705882353,
|
| 76 |
+
"grad_norm": 52.5,
|
| 77 |
+
"learning_rate": 8.823529411764706e-05,
|
| 78 |
+
"loss": 14.5169,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.49019607843137253,
|
| 83 |
+
"grad_norm": 40.25,
|
| 84 |
+
"learning_rate": 9.80392156862745e-05,
|
| 85 |
+
"loss": 7.0407,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.5392156862745098,
|
| 90 |
+
"grad_norm": 17.375,
|
| 91 |
+
"learning_rate": 0.00010784313725490196,
|
| 92 |
+
"loss": 2.4366,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.5882352941176471,
|
| 97 |
+
"grad_norm": 3.015625,
|
| 98 |
+
"learning_rate": 0.00011764705882352942,
|
| 99 |
+
"loss": 2.0315,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.6372549019607843,
|
| 104 |
+
"grad_norm": 2.015625,
|
| 105 |
+
"learning_rate": 0.00012745098039215687,
|
| 106 |
+
"loss": 1.7406,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.6862745098039216,
|
| 111 |
+
"grad_norm": 18.125,
|
| 112 |
+
"learning_rate": 0.0001372549019607843,
|
| 113 |
+
"loss": 1.5659,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.7352941176470589,
|
| 118 |
+
"grad_norm": 3.34375,
|
| 119 |
+
"learning_rate": 0.00014705882352941178,
|
| 120 |
+
"loss": 1.5022,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.7843137254901961,
|
| 125 |
+
"grad_norm": 8.75,
|
| 126 |
+
"learning_rate": 0.00015686274509803922,
|
| 127 |
+
"loss": 1.4175,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.8333333333333334,
|
| 132 |
+
"grad_norm": 2.0625,
|
| 133 |
+
"learning_rate": 0.0001666666666666667,
|
| 134 |
+
"loss": 1.3394,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.8823529411764706,
|
| 139 |
+
"grad_norm": 57.75,
|
| 140 |
+
"learning_rate": 0.00017647058823529413,
|
| 141 |
+
"loss": 1.3717,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.9313725490196079,
|
| 146 |
+
"grad_norm": 6.03125,
|
| 147 |
+
"learning_rate": 0.00018627450980392157,
|
| 148 |
+
"loss": 1.3148,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.9803921568627451,
|
| 153 |
+
"grad_norm": 3.671875,
|
| 154 |
+
"learning_rate": 0.000196078431372549,
|
| 155 |
+
"loss": 1.2436,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.0,
|
| 160 |
+
"eval_loss": 2.589672803878784,
|
| 161 |
+
"eval_runtime": 2.0414,
|
| 162 |
+
"eval_samples_per_second": 4.899,
|
| 163 |
+
"eval_steps_per_second": 0.98,
|
| 164 |
+
"step": 102
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.0294117647058822,
|
| 168 |
+
"grad_norm": 5.65625,
|
| 169 |
+
"learning_rate": 0.00019999472984871732,
|
| 170 |
+
"loss": 1.1863,
|
| 171 |
+
"step": 105
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.0784313725490196,
|
| 175 |
+
"grad_norm": 1.59375,
|
| 176 |
+
"learning_rate": 0.00019996252538028507,
|
| 177 |
+
"loss": 1.1127,
|
| 178 |
+
"step": 110
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.1274509803921569,
|
| 182 |
+
"grad_norm": 2.609375,
|
| 183 |
+
"learning_rate": 0.0001999010537226905,
|
| 184 |
+
"loss": 1.0965,
|
| 185 |
+
"step": 115
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.1764705882352942,
|
| 189 |
+
"grad_norm": 3.125,
|
| 190 |
+
"learning_rate": 0.00019981033287370443,
|
| 191 |
+
"loss": 1.0799,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.2254901960784315,
|
| 196 |
+
"grad_norm": 2.21875,
|
| 197 |
+
"learning_rate": 0.00019969038939472315,
|
| 198 |
+
"loss": 1.0995,
|
| 199 |
+
"step": 125
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.2745098039215685,
|
| 203 |
+
"grad_norm": 13.0625,
|
| 204 |
+
"learning_rate": 0.00019954125840299165,
|
| 205 |
+
"loss": 1.1019,
|
| 206 |
+
"step": 130
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.3235294117647058,
|
| 210 |
+
"grad_norm": 6.09375,
|
| 211 |
+
"learning_rate": 0.00019936298356132176,
|
| 212 |
+
"loss": 1.0908,
|
| 213 |
+
"step": 135
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.3725490196078431,
|
| 217 |
+
"grad_norm": 1.328125,
|
| 218 |
+
"learning_rate": 0.00019915561706530883,
|
| 219 |
+
"loss": 1.092,
|
| 220 |
+
"step": 140
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.4215686274509804,
|
| 224 |
+
"grad_norm": 1.1953125,
|
| 225 |
+
"learning_rate": 0.00019891921962804943,
|
| 226 |
+
"loss": 1.0458,
|
| 227 |
+
"step": 145
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.4705882352941178,
|
| 231 |
+
"grad_norm": 2.578125,
|
| 232 |
+
"learning_rate": 0.00019865386046236596,
|
| 233 |
+
"loss": 1.0323,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.5196078431372548,
|
| 238 |
+
"grad_norm": 2.75,
|
| 239 |
+
"learning_rate": 0.0001983596172605423,
|
| 240 |
+
"loss": 1.0194,
|
| 241 |
+
"step": 155
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.5686274509803921,
|
| 245 |
+
"grad_norm": 3.734375,
|
| 246 |
+
"learning_rate": 0.0001980365761715769,
|
| 247 |
+
"loss": 1.0343,
|
| 248 |
+
"step": 160
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.6176470588235294,
|
| 252 |
+
"grad_norm": 3.296875,
|
| 253 |
+
"learning_rate": 0.0001976848317759601,
|
| 254 |
+
"loss": 0.9982,
|
| 255 |
+
"step": 165
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.6666666666666665,
|
| 259 |
+
"grad_norm": 2.765625,
|
| 260 |
+
"learning_rate": 0.00019730448705798239,
|
| 261 |
+
"loss": 0.9835,
|
| 262 |
+
"step": 170
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.715686274509804,
|
| 266 |
+
"grad_norm": 1.4375,
|
| 267 |
+
"learning_rate": 0.00019689565337558288,
|
| 268 |
+
"loss": 0.965,
|
| 269 |
+
"step": 175
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.7647058823529411,
|
| 273 |
+
"grad_norm": 1.5,
|
| 274 |
+
"learning_rate": 0.00019645845042774553,
|
| 275 |
+
"loss": 0.964,
|
| 276 |
+
"step": 180
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.8137254901960784,
|
| 280 |
+
"grad_norm": 0.75390625,
|
| 281 |
+
"learning_rate": 0.0001959930062194534,
|
| 282 |
+
"loss": 0.9605,
|
| 283 |
+
"step": 185
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.8627450980392157,
|
| 287 |
+
"grad_norm": 14.25,
|
| 288 |
+
"learning_rate": 0.00019549945702421144,
|
| 289 |
+
"loss": 0.9499,
|
| 290 |
+
"step": 190
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.9117647058823528,
|
| 294 |
+
"grad_norm": 6.4375,
|
| 295 |
+
"learning_rate": 0.0001949779473441478,
|
| 296 |
+
"loss": 0.9724,
|
| 297 |
+
"step": 195
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.9607843137254903,
|
| 301 |
+
"grad_norm": 1.015625,
|
| 302 |
+
"learning_rate": 0.00019442862986770646,
|
| 303 |
+
"loss": 0.9498,
|
| 304 |
+
"step": 200
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 2.0,
|
| 308 |
+
"eval_loss": 2.437178134918213,
|
| 309 |
+
"eval_runtime": 2.0579,
|
| 310 |
+
"eval_samples_per_second": 4.859,
|
| 311 |
+
"eval_steps_per_second": 0.972,
|
| 312 |
+
"step": 204
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.0098039215686274,
|
| 316 |
+
"grad_norm": 1.0546875,
|
| 317 |
+
"learning_rate": 0.0001938516654249428,
|
| 318 |
+
"loss": 0.9481,
|
| 319 |
+
"step": 205
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.0588235294117645,
|
| 323 |
+
"grad_norm": 1.2265625,
|
| 324 |
+
"learning_rate": 0.00019324722294043558,
|
| 325 |
+
"loss": 0.8375,
|
| 326 |
+
"step": 210
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.107843137254902,
|
| 330 |
+
"grad_norm": 0.8828125,
|
| 331 |
+
"learning_rate": 0.0001926154793838288,
|
| 332 |
+
"loss": 0.857,
|
| 333 |
+
"step": 215
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.156862745098039,
|
| 337 |
+
"grad_norm": 8.3125,
|
| 338 |
+
"learning_rate": 0.00019195661971801827,
|
| 339 |
+
"loss": 0.8319,
|
| 340 |
+
"step": 220
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.2058823529411766,
|
| 344 |
+
"grad_norm": 1.4609375,
|
| 345 |
+
"learning_rate": 0.00019127083684499806,
|
| 346 |
+
"loss": 0.8394,
|
| 347 |
+
"step": 225
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.2549019607843137,
|
| 351 |
+
"grad_norm": 1.3203125,
|
| 352 |
+
"learning_rate": 0.00019055833154938207,
|
| 353 |
+
"loss": 0.8447,
|
| 354 |
+
"step": 230
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.303921568627451,
|
| 358 |
+
"grad_norm": 1.5078125,
|
| 359 |
+
"learning_rate": 0.00018981931243961824,
|
| 360 |
+
"loss": 0.8494,
|
| 361 |
+
"step": 235
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.3529411764705883,
|
| 365 |
+
"grad_norm": 2.515625,
|
| 366 |
+
"learning_rate": 0.00018905399588691163,
|
| 367 |
+
"loss": 0.8417,
|
| 368 |
+
"step": 240
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.4019607843137254,
|
| 372 |
+
"grad_norm": 0.85546875,
|
| 373 |
+
"learning_rate": 0.00018826260596187505,
|
| 374 |
+
"loss": 0.8389,
|
| 375 |
+
"step": 245
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.450980392156863,
|
| 379 |
+
"grad_norm": 0.98828125,
|
| 380 |
+
"learning_rate": 0.00018744537436892516,
|
| 381 |
+
"loss": 0.8484,
|
| 382 |
+
"step": 250
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.5,
|
| 386 |
+
"grad_norm": 1.03125,
|
| 387 |
+
"learning_rate": 0.00018660254037844388,
|
| 388 |
+
"loss": 0.8427,
|
| 389 |
+
"step": 255
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.549019607843137,
|
| 393 |
+
"grad_norm": 1.28125,
|
| 394 |
+
"learning_rate": 0.00018573435075672424,
|
| 395 |
+
"loss": 0.8546,
|
| 396 |
+
"step": 260
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.5980392156862746,
|
| 400 |
+
"grad_norm": 2.140625,
|
| 401 |
+
"learning_rate": 0.00018484105969372182,
|
| 402 |
+
"loss": 0.8444,
|
| 403 |
+
"step": 265
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.6470588235294117,
|
| 407 |
+
"grad_norm": 1.6328125,
|
| 408 |
+
"learning_rate": 0.00018392292872863267,
|
| 409 |
+
"loss": 0.8594,
|
| 410 |
+
"step": 270
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.696078431372549,
|
| 414 |
+
"grad_norm": 1.3984375,
|
| 415 |
+
"learning_rate": 0.0001829802266733193,
|
| 416 |
+
"loss": 0.821,
|
| 417 |
+
"step": 275
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.7450980392156863,
|
| 421 |
+
"grad_norm": 1.4609375,
|
| 422 |
+
"learning_rate": 0.00018201322953360758,
|
| 423 |
+
"loss": 0.8416,
|
| 424 |
+
"step": 280
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.7941176470588234,
|
| 428 |
+
"grad_norm": 1.1640625,
|
| 429 |
+
"learning_rate": 0.00018102222042847737,
|
| 430 |
+
"loss": 0.8349,
|
| 431 |
+
"step": 285
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.843137254901961,
|
| 435 |
+
"grad_norm": 1.9921875,
|
| 436 |
+
"learning_rate": 0.00018000748950717038,
|
| 437 |
+
"loss": 0.8316,
|
| 438 |
+
"step": 290
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.892156862745098,
|
| 442 |
+
"grad_norm": 3.46875,
|
| 443 |
+
"learning_rate": 0.00017896933386423998,
|
| 444 |
+
"loss": 0.8258,
|
| 445 |
+
"step": 295
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.9411764705882355,
|
| 449 |
+
"grad_norm": 1.0078125,
|
| 450 |
+
"learning_rate": 0.00017790805745256704,
|
| 451 |
+
"loss": 0.8333,
|
| 452 |
+
"step": 300
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.9901960784313726,
|
| 456 |
+
"grad_norm": 5.65625,
|
| 457 |
+
"learning_rate": 0.0001768239709943686,
|
| 458 |
+
"loss": 0.8227,
|
| 459 |
+
"step": 305
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 3.0,
|
| 463 |
+
"eval_loss": 2.4781575202941895,
|
| 464 |
+
"eval_runtime": 2.0485,
|
| 465 |
+
"eval_samples_per_second": 4.882,
|
| 466 |
+
"eval_steps_per_second": 0.976,
|
| 467 |
+
"step": 306
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 3.0392156862745097,
|
| 471 |
+
"grad_norm": 1.015625,
|
| 472 |
+
"learning_rate": 0.00017571739189022365,
|
| 473 |
+
"loss": 0.7236,
|
| 474 |
+
"step": 310
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 3.088235294117647,
|
| 478 |
+
"grad_norm": 0.765625,
|
| 479 |
+
"learning_rate": 0.00017458864412614434,
|
| 480 |
+
"loss": 0.6792,
|
| 481 |
+
"step": 315
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 3.1372549019607843,
|
| 485 |
+
"grad_norm": 0.890625,
|
| 486 |
+
"learning_rate": 0.00017343805817871886,
|
| 487 |
+
"loss": 0.7038,
|
| 488 |
+
"step": 320
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 3.186274509803922,
|
| 492 |
+
"grad_norm": 0.93359375,
|
| 493 |
+
"learning_rate": 0.00017226597091835378,
|
| 494 |
+
"loss": 0.7027,
|
| 495 |
+
"step": 325
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 3.235294117647059,
|
| 499 |
+
"grad_norm": 0.99609375,
|
| 500 |
+
"learning_rate": 0.00017107272551064473,
|
| 501 |
+
"loss": 0.7068,
|
| 502 |
+
"step": 330
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 3.284313725490196,
|
| 506 |
+
"grad_norm": 0.73828125,
|
| 507 |
+
"learning_rate": 0.00016985867131590383,
|
| 508 |
+
"loss": 0.7058,
|
| 509 |
+
"step": 335
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 3.3333333333333335,
|
| 513 |
+
"grad_norm": 1.3515625,
|
| 514 |
+
"learning_rate": 0.0001686241637868734,
|
| 515 |
+
"loss": 0.701,
|
| 516 |
+
"step": 340
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 3.3823529411764706,
|
| 520 |
+
"grad_norm": 2.71875,
|
| 521 |
+
"learning_rate": 0.00016736956436465573,
|
| 522 |
+
"loss": 0.7373,
|
| 523 |
+
"step": 345
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 3.431372549019608,
|
| 527 |
+
"grad_norm": 0.96875,
|
| 528 |
+
"learning_rate": 0.00016609524037289019,
|
| 529 |
+
"loss": 0.7155,
|
| 530 |
+
"step": 350
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 3.480392156862745,
|
| 534 |
+
"grad_norm": 1.1875,
|
| 535 |
+
"learning_rate": 0.00016480156491020727,
|
| 536 |
+
"loss": 0.7245,
|
| 537 |
+
"step": 355
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.5294117647058822,
|
| 541 |
+
"grad_norm": 0.91015625,
|
| 542 |
+
"learning_rate": 0.0001634889167409923,
|
| 543 |
+
"loss": 0.7056,
|
| 544 |
+
"step": 360
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 3.5784313725490198,
|
| 548 |
+
"grad_norm": 1.0546875,
|
| 549 |
+
"learning_rate": 0.00016215768018449012,
|
| 550 |
+
"loss": 0.7182,
|
| 551 |
+
"step": 365
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 3.627450980392157,
|
| 555 |
+
"grad_norm": 0.9140625,
|
| 556 |
+
"learning_rate": 0.00016080824500228367,
|
| 557 |
+
"loss": 0.7209,
|
| 558 |
+
"step": 370
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 3.6764705882352944,
|
| 562 |
+
"grad_norm": 0.91796875,
|
| 563 |
+
"learning_rate": 0.00015944100628417868,
|
| 564 |
+
"loss": 0.7125,
|
| 565 |
+
"step": 375
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 3.7254901960784315,
|
| 569 |
+
"grad_norm": 1.0859375,
|
| 570 |
+
"learning_rate": 0.00015805636433252891,
|
| 571 |
+
"loss": 0.7188,
|
| 572 |
+
"step": 380
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 3.7745098039215685,
|
| 576 |
+
"grad_norm": 0.91015625,
|
| 577 |
+
"learning_rate": 0.00015665472454503483,
|
| 578 |
+
"loss": 0.7289,
|
| 579 |
+
"step": 385
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.8235294117647056,
|
| 583 |
+
"grad_norm": 0.88671875,
|
| 584 |
+
"learning_rate": 0.0001552364972960506,
|
| 585 |
+
"loss": 0.72,
|
| 586 |
+
"step": 390
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 3.872549019607843,
|
| 590 |
+
"grad_norm": 0.76953125,
|
| 591 |
+
"learning_rate": 0.0001538020978164341,
|
| 592 |
+
"loss": 0.731,
|
| 593 |
+
"step": 395
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 3.9215686274509802,
|
| 597 |
+
"grad_norm": 0.91796875,
|
| 598 |
+
"learning_rate": 0.00015235194607197508,
|
| 599 |
+
"loss": 0.7247,
|
| 600 |
+
"step": 400
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 3.9705882352941178,
|
| 604 |
+
"grad_norm": 1.0390625,
|
| 605 |
+
"learning_rate": 0.0001508864666404365,
|
| 606 |
+
"loss": 0.7124,
|
| 607 |
+
"step": 405
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 4.0,
|
| 611 |
+
"eval_loss": 2.6424312591552734,
|
| 612 |
+
"eval_runtime": 2.0434,
|
| 613 |
+
"eval_samples_per_second": 4.894,
|
| 614 |
+
"eval_steps_per_second": 0.979,
|
| 615 |
+
"step": 408
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 4.019607843137255,
|
| 619 |
+
"grad_norm": 1.453125,
|
| 620 |
+
"learning_rate": 0.0001494060885872464,
|
| 621 |
+
"loss": 0.6661,
|
| 622 |
+
"step": 410
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 4.068627450980392,
|
| 626 |
+
"grad_norm": 0.94921875,
|
| 627 |
+
"learning_rate": 0.0001479112453398753,
|
| 628 |
+
"loss": 0.5746,
|
| 629 |
+
"step": 415
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 4.117647058823529,
|
| 633 |
+
"grad_norm": 1.3203125,
|
| 634 |
+
"learning_rate": 0.00014640237456093634,
|
| 635 |
+
"loss": 0.5767,
|
| 636 |
+
"step": 420
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 4.166666666666667,
|
| 640 |
+
"grad_norm": 2.0625,
|
| 641 |
+
"learning_rate": 0.00014487991802004623,
|
| 642 |
+
"loss": 0.566,
|
| 643 |
+
"step": 425
|
| 644 |
+
},
|
| 645 |
+
{
|
| 646 |
+
"epoch": 4.215686274509804,
|
| 647 |
+
"grad_norm": 1.171875,
|
| 648 |
+
"learning_rate": 0.00014334432146448272,
|
| 649 |
+
"loss": 0.5662,
|
| 650 |
+
"step": 430
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 4.264705882352941,
|
| 654 |
+
"grad_norm": 1.0390625,
|
| 655 |
+
"learning_rate": 0.00014179603448867835,
|
| 656 |
+
"loss": 0.5665,
|
| 657 |
+
"step": 435
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 4.313725490196078,
|
| 661 |
+
"grad_norm": 0.84375,
|
| 662 |
+
"learning_rate": 0.00014023551040258725,
|
| 663 |
+
"loss": 0.5675,
|
| 664 |
+
"step": 440
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 4.362745098039215,
|
| 668 |
+
"grad_norm": 0.84765625,
|
| 669 |
+
"learning_rate": 0.00013866320609896447,
|
| 670 |
+
"loss": 0.5763,
|
| 671 |
+
"step": 445
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 4.411764705882353,
|
| 675 |
+
"grad_norm": 0.9140625,
|
| 676 |
+
"learning_rate": 0.00013707958191959608,
|
| 677 |
+
"loss": 0.5709,
|
| 678 |
+
"step": 450
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 4.46078431372549,
|
| 682 |
+
"grad_norm": 0.98046875,
|
| 683 |
+
"learning_rate": 0.00013548510152051963,
|
| 684 |
+
"loss": 0.568,
|
| 685 |
+
"step": 455
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 4.509803921568627,
|
| 689 |
+
"grad_norm": 1.671875,
|
| 690 |
+
"learning_rate": 0.00013388023173627414,
|
| 691 |
+
"loss": 0.57,
|
| 692 |
+
"step": 460
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 4.5588235294117645,
|
| 696 |
+
"grad_norm": 1.0859375,
|
| 697 |
+
"learning_rate": 0.0001322654424432195,
|
| 698 |
+
"loss": 0.5986,
|
| 699 |
+
"step": 465
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 4.607843137254902,
|
| 703 |
+
"grad_norm": 0.8359375,
|
| 704 |
+
"learning_rate": 0.00013064120642196548,
|
| 705 |
+
"loss": 0.5748,
|
| 706 |
+
"step": 470
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 4.6568627450980395,
|
| 710 |
+
"grad_norm": 0.77734375,
|
| 711 |
+
"learning_rate": 0.00012900799921895003,
|
| 712 |
+
"loss": 0.6013,
|
| 713 |
+
"step": 475
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 4.705882352941177,
|
| 717 |
+
"grad_norm": 1.0546875,
|
| 718 |
+
"learning_rate": 0.0001273662990072083,
|
| 719 |
+
"loss": 0.5993,
|
| 720 |
+
"step": 480
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 4.754901960784314,
|
| 724 |
+
"grad_norm": 0.8671875,
|
| 725 |
+
"learning_rate": 0.000125716586446372,
|
| 726 |
+
"loss": 0.5981,
|
| 727 |
+
"step": 485
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 4.803921568627451,
|
| 731 |
+
"grad_norm": 1.046875,
|
| 732 |
+
"learning_rate": 0.00012405934454194146,
|
| 733 |
+
"loss": 0.5892,
|
| 734 |
+
"step": 490
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 4.852941176470588,
|
| 738 |
+
"grad_norm": 0.8984375,
|
| 739 |
+
"learning_rate": 0.0001223950585038703,
|
| 740 |
+
"loss": 0.5831,
|
| 741 |
+
"step": 495
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 4.901960784313726,
|
| 745 |
+
"grad_norm": 0.94140625,
|
| 746 |
+
"learning_rate": 0.00012072421560450497,
|
| 747 |
+
"loss": 0.5911,
|
| 748 |
+
"step": 500
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 4.950980392156863,
|
| 752 |
+
"grad_norm": 1.09375,
|
| 753 |
+
"learning_rate": 0.0001190473050359203,
|
| 754 |
+
"loss": 0.6012,
|
| 755 |
+
"step": 505
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 5.0,
|
| 759 |
+
"grad_norm": 0.91015625,
|
| 760 |
+
"learning_rate": 0.00011736481776669306,
|
| 761 |
+
"loss": 0.5905,
|
| 762 |
+
"step": 510
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 5.0,
|
| 766 |
+
"eval_loss": 3.001382350921631,
|
| 767 |
+
"eval_runtime": 2.0438,
|
| 768 |
+
"eval_samples_per_second": 4.893,
|
| 769 |
+
"eval_steps_per_second": 0.979,
|
| 770 |
+
"step": 510
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 5.049019607843137,
|
| 774 |
+
"grad_norm": 1.140625,
|
| 775 |
+
"learning_rate": 0.00011567724639815544,
|
| 776 |
+
"loss": 0.4544,
|
| 777 |
+
"step": 515
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 5.098039215686274,
|
| 781 |
+
"grad_norm": 1.03125,
|
| 782 |
+
"learning_rate": 0.00011398508502017047,
|
| 783 |
+
"loss": 0.454,
|
| 784 |
+
"step": 520
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 5.147058823529412,
|
| 788 |
+
"grad_norm": 0.9921875,
|
| 789 |
+
"learning_rate": 0.00011228882906647142,
|
| 790 |
+
"loss": 0.4535,
|
| 791 |
+
"step": 525
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 5.196078431372549,
|
| 795 |
+
"grad_norm": 1.1171875,
|
| 796 |
+
"learning_rate": 0.00011058897516960816,
|
| 797 |
+
"loss": 0.4666,
|
| 798 |
+
"step": 530
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 5.245098039215686,
|
| 802 |
+
"grad_norm": 1.0078125,
|
| 803 |
+
"learning_rate": 0.00010888602101554202,
|
| 804 |
+
"loss": 0.4614,
|
| 805 |
+
"step": 535
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 5.294117647058823,
|
| 809 |
+
"grad_norm": 2.0625,
|
| 810 |
+
"learning_rate": 0.00010718046519793276,
|
| 811 |
+
"loss": 0.4565,
|
| 812 |
+
"step": 540
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 5.3431372549019605,
|
| 816 |
+
"grad_norm": 0.8671875,
|
| 817 |
+
"learning_rate": 0.0001054728070721593,
|
| 818 |
+
"loss": 0.4712,
|
| 819 |
+
"step": 545
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 5.392156862745098,
|
| 823 |
+
"grad_norm": 1.078125,
|
| 824 |
+
"learning_rate": 0.00010376354660911771,
|
| 825 |
+
"loss": 0.4596,
|
| 826 |
+
"step": 550
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 5.4411764705882355,
|
| 830 |
+
"grad_norm": 1.0078125,
|
| 831 |
+
"learning_rate": 0.00010205318424883905,
|
| 832 |
+
"loss": 0.4739,
|
| 833 |
+
"step": 555
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 5.490196078431373,
|
| 837 |
+
"grad_norm": 1.328125,
|
| 838 |
+
"learning_rate": 0.00010034222075396953,
|
| 839 |
+
"loss": 0.4578,
|
| 840 |
+
"step": 560
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 5.53921568627451,
|
| 844 |
+
"grad_norm": 1.1328125,
|
| 845 |
+
"learning_rate": 9.863115706315652e-05,
|
| 846 |
+
"loss": 0.4643,
|
| 847 |
+
"step": 565
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 5.588235294117647,
|
| 851 |
+
"grad_norm": 0.921875,
|
| 852 |
+
"learning_rate": 9.692049414438299e-05,
|
| 853 |
+
"loss": 0.4681,
|
| 854 |
+
"step": 570
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 5.637254901960784,
|
| 858 |
+
"grad_norm": 1.234375,
|
| 859 |
+
"learning_rate": 9.521073284829315e-05,
|
| 860 |
+
"loss": 0.474,
|
| 861 |
+
"step": 575
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 5.686274509803922,
|
| 865 |
+
"grad_norm": 1.0078125,
|
| 866 |
+
"learning_rate": 9.350237376155269e-05,
|
| 867 |
+
"loss": 0.4733,
|
| 868 |
+
"step": 580
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 5.735294117647059,
|
| 872 |
+
"grad_norm": 1.203125,
|
| 873 |
+
"learning_rate": 9.179591706028626e-05,
|
| 874 |
+
"loss": 0.4677,
|
| 875 |
+
"step": 585
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 5.784313725490196,
|
| 879 |
+
"grad_norm": 1.1953125,
|
| 880 |
+
"learning_rate": 9.00918623636349e-05,
|
| 881 |
+
"loss": 0.4797,
|
| 882 |
+
"step": 590
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 5.833333333333333,
|
| 886 |
+
"grad_norm": 0.88671875,
|
| 887 |
+
"learning_rate": 8.839070858747697e-05,
|
| 888 |
+
"loss": 0.4801,
|
| 889 |
+
"step": 595
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 5.882352941176471,
|
| 893 |
+
"grad_norm": 0.9453125,
|
| 894 |
+
"learning_rate": 8.669295379835467e-05,
|
| 895 |
+
"loss": 0.4777,
|
| 896 |
+
"step": 600
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 5.931372549019608,
|
| 900 |
+
"grad_norm": 0.953125,
|
| 901 |
+
"learning_rate": 8.499909506764914e-05,
|
| 902 |
+
"loss": 0.4801,
|
| 903 |
+
"step": 605
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 5.980392156862745,
|
| 907 |
+
"grad_norm": 0.9296875,
|
| 908 |
+
"learning_rate": 8.330962832604747e-05,
|
| 909 |
+
"loss": 0.4741,
|
| 910 |
+
"step": 610
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 6.0,
|
| 914 |
+
"eval_loss": 3.4287140369415283,
|
| 915 |
+
"eval_runtime": 2.0461,
|
| 916 |
+
"eval_samples_per_second": 4.887,
|
| 917 |
+
"eval_steps_per_second": 0.977,
|
| 918 |
+
"step": 612
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 6.029411764705882,
|
| 922 |
+
"grad_norm": 0.98046875,
|
| 923 |
+
"learning_rate": 8.162504821834295e-05,
|
| 924 |
+
"loss": 0.4033,
|
| 925 |
+
"step": 615
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 6.078431372549019,
|
| 929 |
+
"grad_norm": 1.0234375,
|
| 930 |
+
"learning_rate": 7.994584795861249e-05,
|
| 931 |
+
"loss": 0.3604,
|
| 932 |
+
"step": 620
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 6.127450980392156,
|
| 936 |
+
"grad_norm": 0.88671875,
|
| 937 |
+
"learning_rate": 7.827251918581226e-05,
|
| 938 |
+
"loss": 0.3601,
|
| 939 |
+
"step": 625
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 6.176470588235294,
|
| 943 |
+
"grad_norm": 0.98828125,
|
| 944 |
+
"learning_rate": 7.660555181983518e-05,
|
| 945 |
+
"loss": 0.3687,
|
| 946 |
+
"step": 630
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 6.2254901960784315,
|
| 950 |
+
"grad_norm": 0.9375,
|
| 951 |
+
"learning_rate": 7.494543391807112e-05,
|
| 952 |
+
"loss": 0.3568,
|
| 953 |
+
"step": 635
|
| 954 |
+
},
|
| 955 |
+
{
|
| 956 |
+
"epoch": 6.2745098039215685,
|
| 957 |
+
"grad_norm": 0.85546875,
|
| 958 |
+
"learning_rate": 7.329265153251285e-05,
|
| 959 |
+
"loss": 0.3726,
|
| 960 |
+
"step": 640
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 6.323529411764706,
|
| 964 |
+
"grad_norm": 0.9453125,
|
| 965 |
+
"learning_rate": 7.164768856744892e-05,
|
| 966 |
+
"loss": 0.3691,
|
| 967 |
+
"step": 645
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 6.372549019607844,
|
| 971 |
+
"grad_norm": 0.9296875,
|
| 972 |
+
"learning_rate": 7.001102663778533e-05,
|
| 973 |
+
"loss": 0.3707,
|
| 974 |
+
"step": 650
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 6.421568627450981,
|
| 978 |
+
"grad_norm": 0.90625,
|
| 979 |
+
"learning_rate": 6.83831449280375e-05,
|
| 980 |
+
"loss": 0.368,
|
| 981 |
+
"step": 655
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 6.470588235294118,
|
| 985 |
+
"grad_norm": 0.859375,
|
| 986 |
+
"learning_rate": 6.676452005203406e-05,
|
| 987 |
+
"loss": 0.3696,
|
| 988 |
+
"step": 660
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 6.519607843137255,
|
| 992 |
+
"grad_norm": 0.9765625,
|
| 993 |
+
"learning_rate": 6.515562591337279e-05,
|
| 994 |
+
"loss": 0.3797,
|
| 995 |
+
"step": 665
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 6.568627450980392,
|
| 999 |
+
"grad_norm": 0.953125,
|
| 1000 |
+
"learning_rate": 6.355693356667065e-05,
|
| 1001 |
+
"loss": 0.3738,
|
| 1002 |
+
"step": 670
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 6.617647058823529,
|
| 1006 |
+
"grad_norm": 0.95703125,
|
| 1007 |
+
"learning_rate": 6.196891107964744e-05,
|
| 1008 |
+
"loss": 0.3747,
|
| 1009 |
+
"step": 675
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 6.666666666666667,
|
| 1013 |
+
"grad_norm": 0.96875,
|
| 1014 |
+
"learning_rate": 6.039202339608432e-05,
|
| 1015 |
+
"loss": 0.3744,
|
| 1016 |
+
"step": 680
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 6.715686274509804,
|
| 1020 |
+
"grad_norm": 0.92578125,
|
| 1021 |
+
"learning_rate": 5.882673219969673e-05,
|
| 1022 |
+
"loss": 0.3701,
|
| 1023 |
+
"step": 685
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 6.764705882352941,
|
| 1027 |
+
"grad_norm": 1.0859375,
|
| 1028 |
+
"learning_rate": 5.727349577896194e-05,
|
| 1029 |
+
"loss": 0.3718,
|
| 1030 |
+
"step": 690
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 6.813725490196078,
|
| 1034 |
+
"grad_norm": 0.859375,
|
| 1035 |
+
"learning_rate": 5.57327688929405e-05,
|
| 1036 |
+
"loss": 0.3721,
|
| 1037 |
+
"step": 695
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 6.862745098039216,
|
| 1041 |
+
"grad_norm": 0.93359375,
|
| 1042 |
+
"learning_rate": 5.4205002638131406e-05,
|
| 1043 |
+
"loss": 0.3774,
|
| 1044 |
+
"step": 700
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 6.911764705882353,
|
| 1048 |
+
"grad_norm": 0.9375,
|
| 1049 |
+
"learning_rate": 5.269064431639901e-05,
|
| 1050 |
+
"loss": 0.3711,
|
| 1051 |
+
"step": 705
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 6.96078431372549,
|
| 1055 |
+
"grad_norm": 0.9375,
|
| 1056 |
+
"learning_rate": 5.119013730401152e-05,
|
| 1057 |
+
"loss": 0.367,
|
| 1058 |
+
"step": 710
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 7.0,
|
| 1062 |
+
"eval_loss": 3.9945197105407715,
|
| 1063 |
+
"eval_runtime": 2.0463,
|
| 1064 |
+
"eval_samples_per_second": 4.887,
|
| 1065 |
+
"eval_steps_per_second": 0.977,
|
| 1066 |
+
"step": 714
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 7.009803921568627,
|
| 1070 |
+
"grad_norm": 0.7734375,
|
| 1071 |
+
"learning_rate": 4.970392092182853e-05,
|
| 1072 |
+
"loss": 0.3551,
|
| 1073 |
+
"step": 715
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 7.0588235294117645,
|
| 1077 |
+
"grad_norm": 0.94140625,
|
| 1078 |
+
"learning_rate": 4.823243030667576e-05,
|
| 1079 |
+
"loss": 0.3014,
|
| 1080 |
+
"step": 720
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 7.107843137254902,
|
| 1084 |
+
"grad_norm": 0.9921875,
|
| 1085 |
+
"learning_rate": 4.6776096283945295e-05,
|
| 1086 |
+
"loss": 0.3016,
|
| 1087 |
+
"step": 725
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 7.1568627450980395,
|
| 1091 |
+
"grad_norm": 0.80078125,
|
| 1092 |
+
"learning_rate": 4.533534524145756e-05,
|
| 1093 |
+
"loss": 0.3027,
|
| 1094 |
+
"step": 730
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 7.205882352941177,
|
| 1098 |
+
"grad_norm": 1.109375,
|
| 1099 |
+
"learning_rate": 4.391059900462304e-05,
|
| 1100 |
+
"loss": 0.2896,
|
| 1101 |
+
"step": 735
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 7.254901960784314,
|
| 1105 |
+
"grad_norm": 0.84765625,
|
| 1106 |
+
"learning_rate": 4.250227471293935e-05,
|
| 1107 |
+
"loss": 0.3093,
|
| 1108 |
+
"step": 740
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 7.303921568627451,
|
| 1112 |
+
"grad_norm": 0.78515625,
|
| 1113 |
+
"learning_rate": 4.1110784697860615e-05,
|
| 1114 |
+
"loss": 0.293,
|
| 1115 |
+
"step": 745
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 7.352941176470588,
|
| 1119 |
+
"grad_norm": 0.921875,
|
| 1120 |
+
"learning_rate": 3.973653636207437e-05,
|
| 1121 |
+
"loss": 0.2891,
|
| 1122 |
+
"step": 750
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 7.401960784313726,
|
| 1126 |
+
"grad_norm": 0.87890625,
|
| 1127 |
+
"learning_rate": 3.8379932060221457e-05,
|
| 1128 |
+
"loss": 0.2968,
|
| 1129 |
+
"step": 755
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 7.450980392156863,
|
| 1133 |
+
"grad_norm": 0.8984375,
|
| 1134 |
+
"learning_rate": 3.7041368981094024e-05,
|
| 1135 |
+
"loss": 0.3001,
|
| 1136 |
+
"step": 760
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 7.5,
|
| 1140 |
+
"grad_norm": 0.84765625,
|
| 1141 |
+
"learning_rate": 3.5721239031346066e-05,
|
| 1142 |
+
"loss": 0.3038,
|
| 1143 |
+
"step": 765
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 7.549019607843137,
|
| 1147 |
+
"grad_norm": 0.859375,
|
| 1148 |
+
"learning_rate": 3.441992872075027e-05,
|
| 1149 |
+
"loss": 0.3035,
|
| 1150 |
+
"step": 770
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 7.598039215686274,
|
| 1154 |
+
"grad_norm": 0.8359375,
|
| 1155 |
+
"learning_rate": 3.313781904903496e-05,
|
| 1156 |
+
"loss": 0.297,
|
| 1157 |
+
"step": 775
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 7.647058823529412,
|
| 1161 |
+
"grad_norm": 0.921875,
|
| 1162 |
+
"learning_rate": 3.187528539433458e-05,
|
| 1163 |
+
"loss": 0.2981,
|
| 1164 |
+
"step": 780
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 7.696078431372549,
|
| 1168 |
+
"grad_norm": 0.8359375,
|
| 1169 |
+
"learning_rate": 3.063269740328579e-05,
|
| 1170 |
+
"loss": 0.3057,
|
| 1171 |
+
"step": 785
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 7.745098039215686,
|
| 1175 |
+
"grad_norm": 0.80859375,
|
| 1176 |
+
"learning_rate": 2.9410418882801683e-05,
|
| 1177 |
+
"loss": 0.2978,
|
| 1178 |
+
"step": 790
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 7.794117647058823,
|
| 1182 |
+
"grad_norm": 0.9140625,
|
| 1183 |
+
"learning_rate": 2.8208807693555818e-05,
|
| 1184 |
+
"loss": 0.2986,
|
| 1185 |
+
"step": 795
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 7.8431372549019605,
|
| 1189 |
+
"grad_norm": 0.94921875,
|
| 1190 |
+
"learning_rate": 2.702821564520732e-05,
|
| 1191 |
+
"loss": 0.3072,
|
| 1192 |
+
"step": 800
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 7.892156862745098,
|
| 1196 |
+
"grad_norm": 0.85546875,
|
| 1197 |
+
"learning_rate": 2.5868988393397374e-05,
|
| 1198 |
+
"loss": 0.3014,
|
| 1199 |
+
"step": 805
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 7.9411764705882355,
|
| 1203 |
+
"grad_norm": 0.8125,
|
| 1204 |
+
"learning_rate": 2.4731465338547556e-05,
|
| 1205 |
+
"loss": 0.2997,
|
| 1206 |
+
"step": 810
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 7.990196078431373,
|
| 1210 |
+
"grad_norm": 0.87890625,
|
| 1211 |
+
"learning_rate": 2.3615979526489772e-05,
|
| 1212 |
+
"loss": 0.3034,
|
| 1213 |
+
"step": 815
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 8.0,
|
| 1217 |
+
"eval_loss": 4.544708728790283,
|
| 1218 |
+
"eval_runtime": 2.0513,
|
| 1219 |
+
"eval_samples_per_second": 4.875,
|
| 1220 |
+
"eval_steps_per_second": 0.975,
|
| 1221 |
+
"step": 816
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 8.03921568627451,
|
| 1225 |
+
"grad_norm": 0.75390625,
|
| 1226 |
+
"learning_rate": 2.252285755095652e-05,
|
| 1227 |
+
"loss": 0.2757,
|
| 1228 |
+
"step": 820
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 8.088235294117647,
|
| 1232 |
+
"grad_norm": 0.72265625,
|
| 1233 |
+
"learning_rate": 2.1452419457960137e-05,
|
| 1234 |
+
"loss": 0.2628,
|
| 1235 |
+
"step": 825
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 8.137254901960784,
|
| 1239 |
+
"grad_norm": 0.921875,
|
| 1240 |
+
"learning_rate": 2.0404978652089323e-05,
|
| 1241 |
+
"loss": 0.2637,
|
| 1242 |
+
"step": 830
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 8.186274509803921,
|
| 1246 |
+
"grad_norm": 0.76953125,
|
| 1247 |
+
"learning_rate": 1.938084180475006e-05,
|
| 1248 |
+
"loss": 0.2664,
|
| 1249 |
+
"step": 835
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 8.235294117647058,
|
| 1253 |
+
"grad_norm": 0.8203125,
|
| 1254 |
+
"learning_rate": 1.8380308764377842e-05,
|
| 1255 |
+
"loss": 0.2658,
|
| 1256 |
+
"step": 840
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 8.284313725490197,
|
| 1260 |
+
"grad_norm": 0.73828125,
|
| 1261 |
+
"learning_rate": 1.7403672468647437e-05,
|
| 1262 |
+
"loss": 0.2697,
|
| 1263 |
+
"step": 845
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 8.333333333333334,
|
| 1267 |
+
"grad_norm": 0.796875,
|
| 1268 |
+
"learning_rate": 1.6451218858706374e-05,
|
| 1269 |
+
"loss": 0.2661,
|
| 1270 |
+
"step": 850
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 8.382352941176471,
|
| 1274 |
+
"grad_norm": 0.703125,
|
| 1275 |
+
"learning_rate": 1.5523226795456347e-05,
|
| 1276 |
+
"loss": 0.2672,
|
| 1277 |
+
"step": 855
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 8.431372549019608,
|
| 1281 |
+
"grad_norm": 0.87109375,
|
| 1282 |
+
"learning_rate": 1.4619967977908156e-05,
|
| 1283 |
+
"loss": 0.2636,
|
| 1284 |
+
"step": 860
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 8.480392156862745,
|
| 1288 |
+
"grad_norm": 0.796875,
|
| 1289 |
+
"learning_rate": 1.3741706863632974e-05,
|
| 1290 |
+
"loss": 0.2679,
|
| 1291 |
+
"step": 865
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 8.529411764705882,
|
| 1295 |
+
"grad_norm": 0.80859375,
|
| 1296 |
+
"learning_rate": 1.2888700591334223e-05,
|
| 1297 |
+
"loss": 0.2712,
|
| 1298 |
+
"step": 870
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 8.57843137254902,
|
| 1302 |
+
"grad_norm": 0.8046875,
|
| 1303 |
+
"learning_rate": 1.2061198905562043e-05,
|
| 1304 |
+
"loss": 0.2667,
|
| 1305 |
+
"step": 875
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 8.627450980392156,
|
| 1309 |
+
"grad_norm": 0.79296875,
|
| 1310 |
+
"learning_rate": 1.1259444083592585e-05,
|
| 1311 |
+
"loss": 0.2655,
|
| 1312 |
+
"step": 880
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 8.676470588235293,
|
| 1316 |
+
"grad_norm": 0.85546875,
|
| 1317 |
+
"learning_rate": 1.0483670864493778e-05,
|
| 1318 |
+
"loss": 0.2631,
|
| 1319 |
+
"step": 885
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 8.72549019607843,
|
| 1323 |
+
"grad_norm": 0.74609375,
|
| 1324 |
+
"learning_rate": 9.734106380398023e-06,
|
| 1325 |
+
"loss": 0.266,
|
| 1326 |
+
"step": 890
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 8.77450980392157,
|
| 1330 |
+
"grad_norm": 0.73046875,
|
| 1331 |
+
"learning_rate": 9.010970090002135e-06,
|
| 1332 |
+
"loss": 0.2593,
|
| 1333 |
+
"step": 895
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 8.823529411764707,
|
| 1337 |
+
"grad_norm": 0.8359375,
|
| 1338 |
+
"learning_rate": 8.314473714313719e-06,
|
| 1339 |
+
"loss": 0.2613,
|
| 1340 |
+
"step": 900
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 8.872549019607844,
|
| 1344 |
+
"grad_norm": 0.75390625,
|
| 1345 |
+
"learning_rate": 7.644821174663307e-06,
|
| 1346 |
+
"loss": 0.2685,
|
| 1347 |
+
"step": 905
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 8.92156862745098,
|
| 1351 |
+
"grad_norm": 0.8046875,
|
| 1352 |
+
"learning_rate": 7.002208532999932e-06,
|
| 1353 |
+
"loss": 0.2684,
|
| 1354 |
+
"step": 910
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 8.970588235294118,
|
| 1358 |
+
"grad_norm": 0.76953125,
|
| 1359 |
+
"learning_rate": 6.386823934487618e-06,
|
| 1360 |
+
"loss": 0.265,
|
| 1361 |
+
"step": 915
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 9.0,
|
| 1365 |
+
"eval_loss": 4.8719377517700195,
|
| 1366 |
+
"eval_runtime": 2.0448,
|
| 1367 |
+
"eval_samples_per_second": 4.89,
|
| 1368 |
+
"eval_steps_per_second": 0.978,
|
| 1369 |
+
"step": 918
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 9.019607843137255,
|
| 1373 |
+
"grad_norm": 0.7421875,
|
| 1374 |
+
"learning_rate": 5.798847552420183e-06,
|
| 1375 |
+
"loss": 0.2551,
|
| 1376 |
+
"step": 920
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 9.068627450980392,
|
| 1380 |
+
"grad_norm": 0.68359375,
|
| 1381 |
+
"learning_rate": 5.238451535469657e-06,
|
| 1382 |
+
"loss": 0.2544,
|
| 1383 |
+
"step": 925
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 9.117647058823529,
|
| 1387 |
+
"grad_norm": 0.81640625,
|
| 1388 |
+
"learning_rate": 4.705799957284351e-06,
|
| 1389 |
+
"loss": 0.2635,
|
| 1390 |
+
"step": 930
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 9.166666666666666,
|
| 1394 |
+
"grad_norm": 0.70703125,
|
| 1395 |
+
"learning_rate": 4.20104876845111e-06,
|
| 1396 |
+
"loss": 0.2586,
|
| 1397 |
+
"step": 935
|
| 1398 |
+
},
|
| 1399 |
+
{
|
| 1400 |
+
"epoch": 9.215686274509803,
|
| 1401 |
+
"grad_norm": 0.7265625,
|
| 1402 |
+
"learning_rate": 3.724345750835878e-06,
|
| 1403 |
+
"loss": 0.2593,
|
| 1404 |
+
"step": 940
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 9.264705882352942,
|
| 1408 |
+
"grad_norm": 0.73828125,
|
| 1409 |
+
"learning_rate": 3.275830474315855e-06,
|
| 1410 |
+
"loss": 0.2631,
|
| 1411 |
+
"step": 945
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 9.313725490196079,
|
| 1415 |
+
"grad_norm": 0.7109375,
|
| 1416 |
+
"learning_rate": 2.855634255915951e-06,
|
| 1417 |
+
"loss": 0.2541,
|
| 1418 |
+
"step": 950
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 9.362745098039216,
|
| 1422 |
+
"grad_norm": 0.83203125,
|
| 1423 |
+
"learning_rate": 2.463880121361717e-06,
|
| 1424 |
+
"loss": 0.2595,
|
| 1425 |
+
"step": 955
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 9.411764705882353,
|
| 1429 |
+
"grad_norm": 0.703125,
|
| 1430 |
+
"learning_rate": 2.100682769059548e-06,
|
| 1431 |
+
"loss": 0.2586,
|
| 1432 |
+
"step": 960
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 9.46078431372549,
|
| 1436 |
+
"grad_norm": 0.71875,
|
| 1437 |
+
"learning_rate": 1.7661485365151553e-06,
|
| 1438 |
+
"loss": 0.2569,
|
| 1439 |
+
"step": 965
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 9.509803921568627,
|
| 1443 |
+
"grad_norm": 0.734375,
|
| 1444 |
+
"learning_rate": 1.4603753691998735e-06,
|
| 1445 |
+
"loss": 0.257,
|
| 1446 |
+
"step": 970
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 9.558823529411764,
|
| 1450 |
+
"grad_norm": 0.81640625,
|
| 1451 |
+
"learning_rate": 1.1834527918740623e-06,
|
| 1452 |
+
"loss": 0.2555,
|
| 1453 |
+
"step": 975
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 9.607843137254902,
|
| 1457 |
+
"grad_norm": 0.73046875,
|
| 1458 |
+
"learning_rate": 9.354618823758655e-07,
|
| 1459 |
+
"loss": 0.2583,
|
| 1460 |
+
"step": 980
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 9.656862745098039,
|
| 1464 |
+
"grad_norm": 0.765625,
|
| 1465 |
+
"learning_rate": 7.164752478831305e-07,
|
| 1466 |
+
"loss": 0.2605,
|
| 1467 |
+
"step": 985
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 9.705882352941176,
|
| 1471 |
+
"grad_norm": 0.74609375,
|
| 1472 |
+
"learning_rate": 5.265570036553813e-07,
|
| 1473 |
+
"loss": 0.2587,
|
| 1474 |
+
"step": 990
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 9.754901960784313,
|
| 1478 |
+
"grad_norm": 0.86328125,
|
| 1479 |
+
"learning_rate": 3.6576275426200015e-07,
|
| 1480 |
+
"loss": 0.2526,
|
| 1481 |
+
"step": 995
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 9.803921568627452,
|
| 1485 |
+
"grad_norm": 0.72265625,
|
| 1486 |
+
"learning_rate": 2.3413957730226144e-07,
|
| 1487 |
+
"loss": 0.2646,
|
| 1488 |
+
"step": 1000
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 9.852941176470589,
|
| 1492 |
+
"grad_norm": 0.73046875,
|
| 1493 |
+
"learning_rate": 1.3172600962190197e-07,
|
| 1494 |
+
"loss": 0.2625,
|
| 1495 |
+
"step": 1005
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 9.901960784313726,
|
| 1499 |
+
"grad_norm": 0.7265625,
|
| 1500 |
+
"learning_rate": 5.8552036030179445e-08,
|
| 1501 |
+
"loss": 0.2573,
|
| 1502 |
+
"step": 1010
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 9.950980392156863,
|
| 1506 |
+
"grad_norm": 0.76953125,
|
| 1507 |
+
"learning_rate": 1.4639080520939542e-08,
|
| 1508 |
+
"loss": 0.2576,
|
| 1509 |
+
"step": 1015
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 10.0,
|
| 1513 |
+
"grad_norm": 0.71875,
|
| 1514 |
+
"learning_rate": 0.0,
|
| 1515 |
+
"loss": 0.2597,
|
| 1516 |
+
"step": 1020
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 10.0,
|
| 1520 |
+
"eval_loss": 4.906526565551758,
|
| 1521 |
+
"eval_runtime": 2.0465,
|
| 1522 |
+
"eval_samples_per_second": 4.886,
|
| 1523 |
+
"eval_steps_per_second": 0.977,
|
| 1524 |
+
"step": 1020
|
| 1525 |
+
},
|
| 1526 |
+
{
|
| 1527 |
+
"epoch": 10.0,
|
| 1528 |
+
"step": 1020,
|
| 1529 |
+
"total_flos": 7.797253181045473e+17,
|
| 1530 |
+
"train_loss": 1.7479132745780197,
|
| 1531 |
+
"train_runtime": 6650.2001,
|
| 1532 |
+
"train_samples_per_second": 2.45,
|
| 1533 |
+
"train_steps_per_second": 0.153
|
| 1534 |
+
}
|
| 1535 |
+
],
|
| 1536 |
+
"logging_steps": 5,
|
| 1537 |
+
"max_steps": 1020,
|
| 1538 |
+
"num_input_tokens_seen": 0,
|
| 1539 |
+
"num_train_epochs": 10,
|
| 1540 |
+
"save_steps": 100,
|
| 1541 |
+
"total_flos": 7.797253181045473e+17,
|
| 1542 |
+
"train_batch_size": 4,
|
| 1543 |
+
"trial_name": null,
|
| 1544 |
+
"trial_params": null
|
| 1545 |
+
}
|