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-2b
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: gemma2b-summarize-gpt4o-32k
|
| 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 |
+
# gemma2b-summarize-gpt4o-32k
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 2.5618
|
| 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: 8
|
| 44 |
+
- eval_batch_size: 8
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 3
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 48
|
| 50 |
+
- total_eval_batch_size: 24
|
| 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 |
+
| 1.4358 | 1.0 | 73 | 2.5470 |
|
| 61 |
+
| 1.2092 | 2.0 | 146 | 2.5064 |
|
| 62 |
+
| 1.1501 | 3.0 | 219 | 2.5000 |
|
| 63 |
+
| 1.0995 | 4.0 | 292 | 2.5064 |
|
| 64 |
+
| 1.0809 | 5.0 | 365 | 2.5177 |
|
| 65 |
+
| 1.0583 | 6.0 | 438 | 2.5419 |
|
| 66 |
+
| 1.04 | 7.0 | 511 | 2.5488 |
|
| 67 |
+
| 1.0248 | 8.0 | 584 | 2.5574 |
|
| 68 |
+
| 1.021 | 9.0 | 657 | 2.5614 |
|
| 69 |
+
| 1.0179 | 10.0 | 730 | 2.5618 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- PEFT 0.11.1
|
| 75 |
+
- Transformers 4.41.2
|
| 76 |
+
- Pytorch 2.3.0+cu121
|
| 77 |
+
- Datasets 2.19.2
|
| 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 19644912
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf020752b7d1b1382c9b71928e7ee85e7e8c494fe1f52c525d9e3eae0aa3c788
|
| 3 |
size 19644912
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.287825372721971e+17,
|
| 4 |
+
"train_loss": 1.1969801510850044,
|
| 5 |
+
"train_runtime": 3891.9077,
|
| 6 |
+
"train_samples": 32305,
|
| 7 |
+
"train_samples_per_second": 9.003,
|
| 8 |
+
"train_steps_per_second": 0.188
|
| 9 |
+
}
|
runs/Jun05_14-45-19_user-HP-Z8-Fury-G5-Workstation-Desktop-PC/events.out.tfevents.1717566340.user-HP-Z8-Fury-G5-Workstation-Desktop-PC.26332.0
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 sha256:43e4fcd25cd566ab9892786c2497c8b3c2721b4c747d9df5153003e068768803
|
| 3 |
+
size 39569
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.287825372721971e+17,
|
| 4 |
+
"train_loss": 1.1969801510850044,
|
| 5 |
+
"train_runtime": 3891.9077,
|
| 6 |
+
"train_samples": 32305,
|
| 7 |
+
"train_samples_per_second": 9.003,
|
| 8 |
+
"train_steps_per_second": 0.188
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1151 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 730,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0136986301369863,
|
| 13 |
+
"grad_norm": 3.5625,
|
| 14 |
+
"learning_rate": 2.7397260273972604e-06,
|
| 15 |
+
"loss": 3.0538,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0684931506849315,
|
| 20 |
+
"grad_norm": 2.296875,
|
| 21 |
+
"learning_rate": 1.3698630136986302e-05,
|
| 22 |
+
"loss": 3.0804,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.136986301369863,
|
| 27 |
+
"grad_norm": 2.59375,
|
| 28 |
+
"learning_rate": 2.7397260273972603e-05,
|
| 29 |
+
"loss": 3.036,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.2054794520547945,
|
| 34 |
+
"grad_norm": 2.1875,
|
| 35 |
+
"learning_rate": 4.1095890410958905e-05,
|
| 36 |
+
"loss": 3.0119,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.273972602739726,
|
| 41 |
+
"grad_norm": 1.84375,
|
| 42 |
+
"learning_rate": 5.479452054794521e-05,
|
| 43 |
+
"loss": 2.7939,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.3424657534246575,
|
| 48 |
+
"grad_norm": 4.4375,
|
| 49 |
+
"learning_rate": 6.84931506849315e-05,
|
| 50 |
+
"loss": 2.5661,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.410958904109589,
|
| 55 |
+
"grad_norm": 7.25,
|
| 56 |
+
"learning_rate": 8.219178082191781e-05,
|
| 57 |
+
"loss": 2.3942,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.4794520547945205,
|
| 62 |
+
"grad_norm": 2.28125,
|
| 63 |
+
"learning_rate": 9.58904109589041e-05,
|
| 64 |
+
"loss": 2.2555,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.547945205479452,
|
| 69 |
+
"grad_norm": 0.94140625,
|
| 70 |
+
"learning_rate": 0.00010958904109589041,
|
| 71 |
+
"loss": 2.1131,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.6164383561643836,
|
| 76 |
+
"grad_norm": 0.6875,
|
| 77 |
+
"learning_rate": 0.0001232876712328767,
|
| 78 |
+
"loss": 1.9336,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.684931506849315,
|
| 83 |
+
"grad_norm": 0.462890625,
|
| 84 |
+
"learning_rate": 0.000136986301369863,
|
| 85 |
+
"loss": 1.7804,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.7534246575342466,
|
| 90 |
+
"grad_norm": 0.39453125,
|
| 91 |
+
"learning_rate": 0.00015068493150684933,
|
| 92 |
+
"loss": 1.6514,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.821917808219178,
|
| 97 |
+
"grad_norm": 0.4453125,
|
| 98 |
+
"learning_rate": 0.00016438356164383562,
|
| 99 |
+
"loss": 1.5682,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.8904109589041096,
|
| 104 |
+
"grad_norm": 0.39453125,
|
| 105 |
+
"learning_rate": 0.00017808219178082192,
|
| 106 |
+
"loss": 1.5013,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.958904109589041,
|
| 111 |
+
"grad_norm": 0.4375,
|
| 112 |
+
"learning_rate": 0.0001917808219178082,
|
| 113 |
+
"loss": 1.4358,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.0,
|
| 118 |
+
"eval_loss": 2.546966075897217,
|
| 119 |
+
"eval_runtime": 0.552,
|
| 120 |
+
"eval_samples_per_second": 18.116,
|
| 121 |
+
"eval_steps_per_second": 1.812,
|
| 122 |
+
"step": 73
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.0273972602739727,
|
| 126 |
+
"grad_norm": 0.2412109375,
|
| 127 |
+
"learning_rate": 0.00019999542705801296,
|
| 128 |
+
"loss": 1.3872,
|
| 129 |
+
"step": 75
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.095890410958904,
|
| 133 |
+
"grad_norm": 0.326171875,
|
| 134 |
+
"learning_rate": 0.00019994398626371643,
|
| 135 |
+
"loss": 1.3485,
|
| 136 |
+
"step": 80
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.1643835616438356,
|
| 140 |
+
"grad_norm": 0.478515625,
|
| 141 |
+
"learning_rate": 0.0001998354179989585,
|
| 142 |
+
"loss": 1.3316,
|
| 143 |
+
"step": 85
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.2328767123287672,
|
| 147 |
+
"grad_norm": 1.1015625,
|
| 148 |
+
"learning_rate": 0.00019966978432080316,
|
| 149 |
+
"loss": 1.3049,
|
| 150 |
+
"step": 90
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.3013698630136985,
|
| 154 |
+
"grad_norm": 0.458984375,
|
| 155 |
+
"learning_rate": 0.00019944717990461207,
|
| 156 |
+
"loss": 1.2894,
|
| 157 |
+
"step": 95
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.36986301369863,
|
| 161 |
+
"grad_norm": 0.427734375,
|
| 162 |
+
"learning_rate": 0.000199167731989929,
|
| 163 |
+
"loss": 1.2703,
|
| 164 |
+
"step": 100
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.4383561643835616,
|
| 168 |
+
"grad_norm": 0.2890625,
|
| 169 |
+
"learning_rate": 0.00019883160030775016,
|
| 170 |
+
"loss": 1.2547,
|
| 171 |
+
"step": 105
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.5068493150684932,
|
| 175 |
+
"grad_norm": 0.291015625,
|
| 176 |
+
"learning_rate": 0.00019843897698922284,
|
| 177 |
+
"loss": 1.2613,
|
| 178 |
+
"step": 110
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.5753424657534247,
|
| 182 |
+
"grad_norm": 0.2734375,
|
| 183 |
+
"learning_rate": 0.0001979900864558242,
|
| 184 |
+
"loss": 1.2388,
|
| 185 |
+
"step": 115
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.643835616438356,
|
| 189 |
+
"grad_norm": 0.275390625,
|
| 190 |
+
"learning_rate": 0.00019748518529108316,
|
| 191 |
+
"loss": 1.2528,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.7123287671232876,
|
| 196 |
+
"grad_norm": 0.419921875,
|
| 197 |
+
"learning_rate": 0.00019692456209391846,
|
| 198 |
+
"loss": 1.2373,
|
| 199 |
+
"step": 125
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.7808219178082192,
|
| 203 |
+
"grad_norm": 0.8046875,
|
| 204 |
+
"learning_rate": 0.00019630853731367713,
|
| 205 |
+
"loss": 1.2241,
|
| 206 |
+
"step": 130
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.8493150684931505,
|
| 210 |
+
"grad_norm": 0.33984375,
|
| 211 |
+
"learning_rate": 0.0001956374630669672,
|
| 212 |
+
"loss": 1.2263,
|
| 213 |
+
"step": 135
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.9178082191780823,
|
| 217 |
+
"grad_norm": 0.287109375,
|
| 218 |
+
"learning_rate": 0.00019491172293638968,
|
| 219 |
+
"loss": 1.212,
|
| 220 |
+
"step": 140
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.9863013698630136,
|
| 224 |
+
"grad_norm": 0.2734375,
|
| 225 |
+
"learning_rate": 0.00019413173175128473,
|
| 226 |
+
"loss": 1.2092,
|
| 227 |
+
"step": 145
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 2.0,
|
| 231 |
+
"eval_loss": 2.5063788890838623,
|
| 232 |
+
"eval_runtime": 0.5512,
|
| 233 |
+
"eval_samples_per_second": 18.144,
|
| 234 |
+
"eval_steps_per_second": 1.814,
|
| 235 |
+
"step": 146
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 2.0547945205479454,
|
| 239 |
+
"grad_norm": 0.380859375,
|
| 240 |
+
"learning_rate": 0.00019329793535061723,
|
| 241 |
+
"loss": 1.1791,
|
| 242 |
+
"step": 150
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.1232876712328768,
|
| 246 |
+
"grad_norm": 0.458984375,
|
| 247 |
+
"learning_rate": 0.00019241081032813772,
|
| 248 |
+
"loss": 1.1745,
|
| 249 |
+
"step": 155
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.191780821917808,
|
| 253 |
+
"grad_norm": 0.380859375,
|
| 254 |
+
"learning_rate": 0.0001914708637599636,
|
| 255 |
+
"loss": 1.1783,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 2.26027397260274,
|
| 260 |
+
"grad_norm": 0.2734375,
|
| 261 |
+
"learning_rate": 0.00019047863291473717,
|
| 262 |
+
"loss": 1.156,
|
| 263 |
+
"step": 165
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.328767123287671,
|
| 267 |
+
"grad_norm": 0.5625,
|
| 268 |
+
"learning_rate": 0.0001894346849465257,
|
| 269 |
+
"loss": 1.1642,
|
| 270 |
+
"step": 170
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 2.3972602739726026,
|
| 274 |
+
"grad_norm": 0.65625,
|
| 275 |
+
"learning_rate": 0.00018833961657063885,
|
| 276 |
+
"loss": 1.176,
|
| 277 |
+
"step": 175
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 2.4657534246575343,
|
| 281 |
+
"grad_norm": 0.640625,
|
| 282 |
+
"learning_rate": 0.00018719405372254948,
|
| 283 |
+
"loss": 1.16,
|
| 284 |
+
"step": 180
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.5342465753424657,
|
| 288 |
+
"grad_norm": 0.400390625,
|
| 289 |
+
"learning_rate": 0.00018599865120011192,
|
| 290 |
+
"loss": 1.1653,
|
| 291 |
+
"step": 185
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 2.602739726027397,
|
| 295 |
+
"grad_norm": 0.3125,
|
| 296 |
+
"learning_rate": 0.00018475409228928312,
|
| 297 |
+
"loss": 1.1616,
|
| 298 |
+
"step": 190
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 2.671232876712329,
|
| 302 |
+
"grad_norm": 0.6796875,
|
| 303 |
+
"learning_rate": 0.00018346108837355972,
|
| 304 |
+
"loss": 1.1549,
|
| 305 |
+
"step": 195
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 2.73972602739726,
|
| 309 |
+
"grad_norm": 0.54296875,
|
| 310 |
+
"learning_rate": 0.00018212037852735486,
|
| 311 |
+
"loss": 1.1518,
|
| 312 |
+
"step": 200
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 2.808219178082192,
|
| 316 |
+
"grad_norm": 0.52734375,
|
| 317 |
+
"learning_rate": 0.00018073272909354727,
|
| 318 |
+
"loss": 1.1516,
|
| 319 |
+
"step": 205
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 2.8767123287671232,
|
| 323 |
+
"grad_norm": 0.375,
|
| 324 |
+
"learning_rate": 0.00017929893324544332,
|
| 325 |
+
"loss": 1.154,
|
| 326 |
+
"step": 210
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.9452054794520546,
|
| 330 |
+
"grad_norm": 0.44921875,
|
| 331 |
+
"learning_rate": 0.00017781981053340337,
|
| 332 |
+
"loss": 1.1501,
|
| 333 |
+
"step": 215
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 3.0,
|
| 337 |
+
"eval_loss": 2.4999959468841553,
|
| 338 |
+
"eval_runtime": 0.5406,
|
| 339 |
+
"eval_samples_per_second": 18.499,
|
| 340 |
+
"eval_steps_per_second": 1.85,
|
| 341 |
+
"step": 219
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 3.0136986301369864,
|
| 345 |
+
"grad_norm": 0.353515625,
|
| 346 |
+
"learning_rate": 0.00017629620641639103,
|
| 347 |
+
"loss": 1.1368,
|
| 348 |
+
"step": 220
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 3.0821917808219177,
|
| 352 |
+
"grad_norm": 0.337890625,
|
| 353 |
+
"learning_rate": 0.00017472899177871297,
|
| 354 |
+
"loss": 1.118,
|
| 355 |
+
"step": 225
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 3.1506849315068495,
|
| 359 |
+
"grad_norm": 0.353515625,
|
| 360 |
+
"learning_rate": 0.00017311906243222614,
|
| 361 |
+
"loss": 1.1122,
|
| 362 |
+
"step": 230
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 3.219178082191781,
|
| 366 |
+
"grad_norm": 0.34765625,
|
| 367 |
+
"learning_rate": 0.00017146733860429612,
|
| 368 |
+
"loss": 1.1205,
|
| 369 |
+
"step": 235
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.287671232876712,
|
| 373 |
+
"grad_norm": 0.5625,
|
| 374 |
+
"learning_rate": 0.00016977476441179992,
|
| 375 |
+
"loss": 1.1197,
|
| 376 |
+
"step": 240
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 3.356164383561644,
|
| 380 |
+
"grad_norm": 0.796875,
|
| 381 |
+
"learning_rate": 0.0001680423073214737,
|
| 382 |
+
"loss": 1.1155,
|
| 383 |
+
"step": 245
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.4246575342465753,
|
| 387 |
+
"grad_norm": 0.47265625,
|
| 388 |
+
"learning_rate": 0.00016627095759691362,
|
| 389 |
+
"loss": 1.1255,
|
| 390 |
+
"step": 250
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 3.493150684931507,
|
| 394 |
+
"grad_norm": 0.451171875,
|
| 395 |
+
"learning_rate": 0.00016446172773254629,
|
| 396 |
+
"loss": 1.1113,
|
| 397 |
+
"step": 255
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 3.5616438356164384,
|
| 401 |
+
"grad_norm": 0.486328125,
|
| 402 |
+
"learning_rate": 0.0001626156518748922,
|
| 403 |
+
"loss": 1.1073,
|
| 404 |
+
"step": 260
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 3.6301369863013697,
|
| 408 |
+
"grad_norm": 1.046875,
|
| 409 |
+
"learning_rate": 0.0001607337852314527,
|
| 410 |
+
"loss": 1.1155,
|
| 411 |
+
"step": 265
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 3.6986301369863015,
|
| 415 |
+
"grad_norm": 0.6171875,
|
| 416 |
+
"learning_rate": 0.00015881720346755905,
|
| 417 |
+
"loss": 1.1239,
|
| 418 |
+
"step": 270
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 3.767123287671233,
|
| 422 |
+
"grad_norm": 0.416015625,
|
| 423 |
+
"learning_rate": 0.00015686700209152738,
|
| 424 |
+
"loss": 1.1185,
|
| 425 |
+
"step": 275
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 3.8356164383561646,
|
| 429 |
+
"grad_norm": 0.380859375,
|
| 430 |
+
"learning_rate": 0.00015488429582847192,
|
| 431 |
+
"loss": 1.1125,
|
| 432 |
+
"step": 280
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 3.904109589041096,
|
| 436 |
+
"grad_norm": 0.51953125,
|
| 437 |
+
"learning_rate": 0.0001528702179831338,
|
| 438 |
+
"loss": 1.0925,
|
| 439 |
+
"step": 285
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 3.9726027397260273,
|
| 443 |
+
"grad_norm": 0.3671875,
|
| 444 |
+
"learning_rate": 0.00015082591979208976,
|
| 445 |
+
"loss": 1.0995,
|
| 446 |
+
"step": 290
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 4.0,
|
| 450 |
+
"eval_loss": 2.506415843963623,
|
| 451 |
+
"eval_runtime": 0.5555,
|
| 452 |
+
"eval_samples_per_second": 18.001,
|
| 453 |
+
"eval_steps_per_second": 1.8,
|
| 454 |
+
"step": 292
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 4.041095890410959,
|
| 458 |
+
"grad_norm": 0.515625,
|
| 459 |
+
"learning_rate": 0.00014875256976571135,
|
| 460 |
+
"loss": 1.0993,
|
| 461 |
+
"step": 295
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 4.109589041095891,
|
| 465 |
+
"grad_norm": 0.375,
|
| 466 |
+
"learning_rate": 0.00014665135302025035,
|
| 467 |
+
"loss": 1.0789,
|
| 468 |
+
"step": 300
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 4.178082191780822,
|
| 472 |
+
"grad_norm": 0.3984375,
|
| 473 |
+
"learning_rate": 0.00014452347060043237,
|
| 474 |
+
"loss": 1.0724,
|
| 475 |
+
"step": 305
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 4.2465753424657535,
|
| 479 |
+
"grad_norm": 0.447265625,
|
| 480 |
+
"learning_rate": 0.0001423701387929459,
|
| 481 |
+
"loss": 1.0874,
|
| 482 |
+
"step": 310
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 4.315068493150685,
|
| 486 |
+
"grad_norm": 0.357421875,
|
| 487 |
+
"learning_rate": 0.00014019258843121893,
|
| 488 |
+
"loss": 1.0913,
|
| 489 |
+
"step": 315
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 4.383561643835616,
|
| 493 |
+
"grad_norm": 0.35546875,
|
| 494 |
+
"learning_rate": 0.00013799206419188103,
|
| 495 |
+
"loss": 1.085,
|
| 496 |
+
"step": 320
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 4.4520547945205475,
|
| 500 |
+
"grad_norm": 0.46875,
|
| 501 |
+
"learning_rate": 0.0001357698238833126,
|
| 502 |
+
"loss": 1.0733,
|
| 503 |
+
"step": 325
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 4.52054794520548,
|
| 507 |
+
"grad_norm": 0.44140625,
|
| 508 |
+
"learning_rate": 0.00013352713772668765,
|
| 509 |
+
"loss": 1.0799,
|
| 510 |
+
"step": 330
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 4.589041095890411,
|
| 514 |
+
"grad_norm": 0.412109375,
|
| 515 |
+
"learning_rate": 0.00013126528762992247,
|
| 516 |
+
"loss": 1.0774,
|
| 517 |
+
"step": 335
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 4.657534246575342,
|
| 521 |
+
"grad_norm": 0.34375,
|
| 522 |
+
"learning_rate": 0.00012898556645494325,
|
| 523 |
+
"loss": 1.0742,
|
| 524 |
+
"step": 340
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 4.726027397260274,
|
| 528 |
+
"grad_norm": 0.408203125,
|
| 529 |
+
"learning_rate": 0.0001266892772786929,
|
| 530 |
+
"loss": 1.0806,
|
| 531 |
+
"step": 345
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 4.794520547945205,
|
| 535 |
+
"grad_norm": 0.38671875,
|
| 536 |
+
"learning_rate": 0.00012437773264829897,
|
| 537 |
+
"loss": 1.0892,
|
| 538 |
+
"step": 350
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 4.863013698630137,
|
| 542 |
+
"grad_norm": 0.482421875,
|
| 543 |
+
"learning_rate": 0.00012205225383082843,
|
| 544 |
+
"loss": 1.0805,
|
| 545 |
+
"step": 355
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 4.931506849315069,
|
| 549 |
+
"grad_norm": 0.3828125,
|
| 550 |
+
"learning_rate": 0.00011971417005805818,
|
| 551 |
+
"loss": 1.0786,
|
| 552 |
+
"step": 360
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 5.0,
|
| 556 |
+
"grad_norm": 0.37890625,
|
| 557 |
+
"learning_rate": 0.00011736481776669306,
|
| 558 |
+
"loss": 1.0809,
|
| 559 |
+
"step": 365
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 5.0,
|
| 563 |
+
"eval_loss": 2.5177359580993652,
|
| 564 |
+
"eval_runtime": 0.545,
|
| 565 |
+
"eval_samples_per_second": 18.35,
|
| 566 |
+
"eval_steps_per_second": 1.835,
|
| 567 |
+
"step": 365
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 5.068493150684931,
|
| 571 |
+
"grad_norm": 0.388671875,
|
| 572 |
+
"learning_rate": 0.00011500553983446527,
|
| 573 |
+
"loss": 1.0503,
|
| 574 |
+
"step": 370
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 5.136986301369863,
|
| 578 |
+
"grad_norm": 0.38671875,
|
| 579 |
+
"learning_rate": 0.00011263768481255264,
|
| 580 |
+
"loss": 1.0578,
|
| 581 |
+
"step": 375
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 5.205479452054795,
|
| 585 |
+
"grad_norm": 0.41015625,
|
| 586 |
+
"learning_rate": 0.00011026260615475333,
|
| 587 |
+
"loss": 1.0601,
|
| 588 |
+
"step": 380
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 5.273972602739726,
|
| 592 |
+
"grad_norm": 0.40625,
|
| 593 |
+
"learning_rate": 0.00010788166144385888,
|
| 594 |
+
"loss": 1.0519,
|
| 595 |
+
"step": 385
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 5.342465753424658,
|
| 599 |
+
"grad_norm": 0.515625,
|
| 600 |
+
"learning_rate": 0.0001054962116156667,
|
| 601 |
+
"loss": 1.0555,
|
| 602 |
+
"step": 390
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 5.410958904109589,
|
| 606 |
+
"grad_norm": 0.56640625,
|
| 607 |
+
"learning_rate": 0.0001031076201810762,
|
| 608 |
+
"loss": 1.0673,
|
| 609 |
+
"step": 395
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 5.47945205479452,
|
| 613 |
+
"grad_norm": 0.48046875,
|
| 614 |
+
"learning_rate": 0.00010071725244671282,
|
| 615 |
+
"loss": 1.0495,
|
| 616 |
+
"step": 400
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 5.5479452054794525,
|
| 620 |
+
"grad_norm": 0.478515625,
|
| 621 |
+
"learning_rate": 9.83264747345259e-05,
|
| 622 |
+
"loss": 1.054,
|
| 623 |
+
"step": 405
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 5.616438356164384,
|
| 627 |
+
"grad_norm": 0.408203125,
|
| 628 |
+
"learning_rate": 9.593665360080599e-05,
|
| 629 |
+
"loss": 1.0631,
|
| 630 |
+
"step": 410
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 5.684931506849315,
|
| 634 |
+
"grad_norm": 0.376953125,
|
| 635 |
+
"learning_rate": 9.354915505506839e-05,
|
| 636 |
+
"loss": 1.0599,
|
| 637 |
+
"step": 415
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 5.7534246575342465,
|
| 641 |
+
"grad_norm": 0.373046875,
|
| 642 |
+
"learning_rate": 9.116534377924883e-05,
|
| 643 |
+
"loss": 1.0554,
|
| 644 |
+
"step": 420
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 5.821917808219178,
|
| 648 |
+
"grad_norm": 0.40625,
|
| 649 |
+
"learning_rate": 8.878658234765858e-05,
|
| 650 |
+
"loss": 1.0503,
|
| 651 |
+
"step": 425
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 5.890410958904109,
|
| 655 |
+
"grad_norm": 0.39453125,
|
| 656 |
+
"learning_rate": 8.641423044814374e-05,
|
| 657 |
+
"loss": 1.0632,
|
| 658 |
+
"step": 430
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 5.958904109589041,
|
| 662 |
+
"grad_norm": 0.431640625,
|
| 663 |
+
"learning_rate": 8.404964410489485e-05,
|
| 664 |
+
"loss": 1.0583,
|
| 665 |
+
"step": 435
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 6.0,
|
| 669 |
+
"eval_loss": 2.541933536529541,
|
| 670 |
+
"eval_runtime": 0.5507,
|
| 671 |
+
"eval_samples_per_second": 18.16,
|
| 672 |
+
"eval_steps_per_second": 1.816,
|
| 673 |
+
"step": 438
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 6.027397260273973,
|
| 677 |
+
"grad_norm": 0.3828125,
|
| 678 |
+
"learning_rate": 8.169417490335007e-05,
|
| 679 |
+
"loss": 1.0504,
|
| 680 |
+
"step": 440
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 6.095890410958904,
|
| 684 |
+
"grad_norm": 0.38671875,
|
| 685 |
+
"learning_rate": 7.934916921763628e-05,
|
| 686 |
+
"loss": 1.0301,
|
| 687 |
+
"step": 445
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 6.164383561643835,
|
| 691 |
+
"grad_norm": 0.3828125,
|
| 692 |
+
"learning_rate": 7.701596744098818e-05,
|
| 693 |
+
"loss": 1.0358,
|
| 694 |
+
"step": 450
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 6.232876712328767,
|
| 698 |
+
"grad_norm": 0.40625,
|
| 699 |
+
"learning_rate": 7.469590321958662e-05,
|
| 700 |
+
"loss": 1.0316,
|
| 701 |
+
"step": 455
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 6.301369863013699,
|
| 705 |
+
"grad_norm": 0.59765625,
|
| 706 |
+
"learning_rate": 7.239030269025311e-05,
|
| 707 |
+
"loss": 1.0415,
|
| 708 |
+
"step": 460
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 6.36986301369863,
|
| 712 |
+
"grad_norm": 0.482421875,
|
| 713 |
+
"learning_rate": 7.010048372243698e-05,
|
| 714 |
+
"loss": 1.0336,
|
| 715 |
+
"step": 465
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 6.438356164383562,
|
| 719 |
+
"grad_norm": 0.40625,
|
| 720 |
+
"learning_rate": 6.782775516492771e-05,
|
| 721 |
+
"loss": 1.0496,
|
| 722 |
+
"step": 470
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 6.506849315068493,
|
| 726 |
+
"grad_norm": 0.38671875,
|
| 727 |
+
"learning_rate": 6.5573416097724e-05,
|
| 728 |
+
"loss": 1.04,
|
| 729 |
+
"step": 475
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 6.575342465753424,
|
| 733 |
+
"grad_norm": 0.46875,
|
| 734 |
+
"learning_rate": 6.333875508948593e-05,
|
| 735 |
+
"loss": 1.0461,
|
| 736 |
+
"step": 480
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 6.6438356164383565,
|
| 740 |
+
"grad_norm": 0.4296875,
|
| 741 |
+
"learning_rate": 6.112504946099604e-05,
|
| 742 |
+
"loss": 1.0439,
|
| 743 |
+
"step": 485
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 6.712328767123288,
|
| 747 |
+
"grad_norm": 0.42578125,
|
| 748 |
+
"learning_rate": 5.8933564555049105e-05,
|
| 749 |
+
"loss": 1.0358,
|
| 750 |
+
"step": 490
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 6.780821917808219,
|
| 754 |
+
"grad_norm": 0.5234375,
|
| 755 |
+
"learning_rate": 5.6765553013188766e-05,
|
| 756 |
+
"loss": 1.045,
|
| 757 |
+
"step": 495
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 6.8493150684931505,
|
| 761 |
+
"grad_norm": 0.365234375,
|
| 762 |
+
"learning_rate": 5.462225405970401e-05,
|
| 763 |
+
"loss": 1.0329,
|
| 764 |
+
"step": 500
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 6.917808219178082,
|
| 768 |
+
"grad_norm": 0.388671875,
|
| 769 |
+
"learning_rate": 5.2504892793295e-05,
|
| 770 |
+
"loss": 1.0442,
|
| 771 |
+
"step": 505
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 6.986301369863014,
|
| 775 |
+
"grad_norm": 0.37109375,
|
| 776 |
+
"learning_rate": 5.041467948681269e-05,
|
| 777 |
+
"loss": 1.04,
|
| 778 |
+
"step": 510
|
| 779 |
+
},
|
| 780 |
+
{
|
| 781 |
+
"epoch": 7.0,
|
| 782 |
+
"eval_loss": 2.5487561225891113,
|
| 783 |
+
"eval_runtime": 0.5536,
|
| 784 |
+
"eval_samples_per_second": 18.064,
|
| 785 |
+
"eval_steps_per_second": 1.806,
|
| 786 |
+
"step": 511
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 7.054794520547945,
|
| 790 |
+
"grad_norm": 0.384765625,
|
| 791 |
+
"learning_rate": 4.835280889547351e-05,
|
| 792 |
+
"loss": 1.0298,
|
| 793 |
+
"step": 515
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 7.123287671232877,
|
| 797 |
+
"grad_norm": 0.376953125,
|
| 798 |
+
"learning_rate": 4.6320459573942856e-05,
|
| 799 |
+
"loss": 1.0213,
|
| 800 |
+
"step": 520
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 7.191780821917808,
|
| 804 |
+
"grad_norm": 0.390625,
|
| 805 |
+
"learning_rate": 4.431879320267972e-05,
|
| 806 |
+
"loss": 1.0334,
|
| 807 |
+
"step": 525
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 7.260273972602739,
|
| 811 |
+
"grad_norm": 0.390625,
|
| 812 |
+
"learning_rate": 4.2348953923925916e-05,
|
| 813 |
+
"loss": 1.0208,
|
| 814 |
+
"step": 530
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 7.328767123287671,
|
| 818 |
+
"grad_norm": 0.365234375,
|
| 819 |
+
"learning_rate": 4.041206768772022e-05,
|
| 820 |
+
"loss": 1.0323,
|
| 821 |
+
"step": 535
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 7.397260273972603,
|
| 825 |
+
"grad_norm": 0.41015625,
|
| 826 |
+
"learning_rate": 3.850924160831115e-05,
|
| 827 |
+
"loss": 1.0295,
|
| 828 |
+
"step": 540
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 7.465753424657534,
|
| 832 |
+
"grad_norm": 0.373046875,
|
| 833 |
+
"learning_rate": 3.6641563331336125e-05,
|
| 834 |
+
"loss": 1.0266,
|
| 835 |
+
"step": 545
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 7.534246575342466,
|
| 839 |
+
"grad_norm": 0.3828125,
|
| 840 |
+
"learning_rate": 3.4810100412128747e-05,
|
| 841 |
+
"loss": 1.0218,
|
| 842 |
+
"step": 550
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 7.602739726027397,
|
| 846 |
+
"grad_norm": 0.375,
|
| 847 |
+
"learning_rate": 3.3015899705509734e-05,
|
| 848 |
+
"loss": 1.0369,
|
| 849 |
+
"step": 555
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 7.671232876712329,
|
| 853 |
+
"grad_norm": 0.40234375,
|
| 854 |
+
"learning_rate": 3.125998676740987e-05,
|
| 855 |
+
"loss": 1.0308,
|
| 856 |
+
"step": 560
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 7.739726027397261,
|
| 860 |
+
"grad_norm": 0.3671875,
|
| 861 |
+
"learning_rate": 2.9543365268667867e-05,
|
| 862 |
+
"loss": 1.0225,
|
| 863 |
+
"step": 565
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 7.808219178082192,
|
| 867 |
+
"grad_norm": 0.36328125,
|
| 868 |
+
"learning_rate": 2.7867016421336776e-05,
|
| 869 |
+
"loss": 1.0236,
|
| 870 |
+
"step": 570
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 7.876712328767123,
|
| 874 |
+
"grad_norm": 0.3828125,
|
| 875 |
+
"learning_rate": 2.6231898417828603e-05,
|
| 876 |
+
"loss": 1.0272,
|
| 877 |
+
"step": 575
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 7.945205479452055,
|
| 881 |
+
"grad_norm": 0.37890625,
|
| 882 |
+
"learning_rate": 2.4638945883216235e-05,
|
| 883 |
+
"loss": 1.0248,
|
| 884 |
+
"step": 580
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 8.0,
|
| 888 |
+
"eval_loss": 2.557366132736206,
|
| 889 |
+
"eval_runtime": 0.554,
|
| 890 |
+
"eval_samples_per_second": 18.051,
|
| 891 |
+
"eval_steps_per_second": 1.805,
|
| 892 |
+
"step": 584
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 8.013698630136986,
|
| 896 |
+
"grad_norm": 0.365234375,
|
| 897 |
+
"learning_rate": 2.3089069341006565e-05,
|
| 898 |
+
"loss": 1.0197,
|
| 899 |
+
"step": 585
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 8.082191780821917,
|
| 903 |
+
"grad_norm": 0.359375,
|
| 904 |
+
"learning_rate": 2.1583154692689976e-05,
|
| 905 |
+
"loss": 1.0234,
|
| 906 |
+
"step": 590
|
| 907 |
+
},
|
| 908 |
+
{
|
| 909 |
+
"epoch": 8.150684931506849,
|
| 910 |
+
"grad_norm": 0.435546875,
|
| 911 |
+
"learning_rate": 2.0122062711363532e-05,
|
| 912 |
+
"loss": 1.0212,
|
| 913 |
+
"step": 595
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 8.219178082191782,
|
| 917 |
+
"grad_norm": 0.357421875,
|
| 918 |
+
"learning_rate": 1.8706628549717452e-05,
|
| 919 |
+
"loss": 1.0168,
|
| 920 |
+
"step": 600
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 8.287671232876713,
|
| 924 |
+
"grad_norm": 0.37109375,
|
| 925 |
+
"learning_rate": 1.7337661262666294e-05,
|
| 926 |
+
"loss": 1.0172,
|
| 927 |
+
"step": 605
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 8.356164383561644,
|
| 931 |
+
"grad_norm": 0.373046875,
|
| 932 |
+
"learning_rate": 1.601594334489702e-05,
|
| 933 |
+
"loss": 1.0167,
|
| 934 |
+
"step": 610
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 8.424657534246576,
|
| 938 |
+
"grad_norm": 0.3828125,
|
| 939 |
+
"learning_rate": 1.474223028359939e-05,
|
| 940 |
+
"loss": 1.0271,
|
| 941 |
+
"step": 615
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 8.493150684931507,
|
| 945 |
+
"grad_norm": 0.369140625,
|
| 946 |
+
"learning_rate": 1.3517250126632986e-05,
|
| 947 |
+
"loss": 1.0233,
|
| 948 |
+
"step": 620
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 8.561643835616438,
|
| 952 |
+
"grad_norm": 0.3671875,
|
| 953 |
+
"learning_rate": 1.2341703066379074e-05,
|
| 954 |
+
"loss": 1.0209,
|
| 955 |
+
"step": 625
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 8.63013698630137,
|
| 959 |
+
"grad_norm": 0.376953125,
|
| 960 |
+
"learning_rate": 1.1216261039514087e-05,
|
| 961 |
+
"loss": 1.0143,
|
| 962 |
+
"step": 630
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 8.698630136986301,
|
| 966 |
+
"grad_norm": 0.380859375,
|
| 967 |
+
"learning_rate": 1.0141567342934132e-05,
|
| 968 |
+
"loss": 1.0231,
|
| 969 |
+
"step": 635
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 8.767123287671232,
|
| 973 |
+
"grad_norm": 0.35546875,
|
| 974 |
+
"learning_rate": 9.118236266049707e-06,
|
| 975 |
+
"loss": 1.014,
|
| 976 |
+
"step": 640
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 8.835616438356164,
|
| 980 |
+
"grad_norm": 0.375,
|
| 981 |
+
"learning_rate": 8.146852739661105e-06,
|
| 982 |
+
"loss": 1.016,
|
| 983 |
+
"step": 645
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 8.904109589041095,
|
| 987 |
+
"grad_norm": 0.373046875,
|
| 988 |
+
"learning_rate": 7.2279720016148244e-06,
|
| 989 |
+
"loss": 1.0274,
|
| 990 |
+
"step": 650
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 8.972602739726028,
|
| 994 |
+
"grad_norm": 0.392578125,
|
| 995 |
+
"learning_rate": 6.36211927943271e-06,
|
| 996 |
+
"loss": 1.021,
|
| 997 |
+
"step": 655
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 9.0,
|
| 1001 |
+
"eval_loss": 2.5613903999328613,
|
| 1002 |
+
"eval_runtime": 0.5421,
|
| 1003 |
+
"eval_samples_per_second": 18.447,
|
| 1004 |
+
"eval_steps_per_second": 1.845,
|
| 1005 |
+
"step": 657
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 9.04109589041096,
|
| 1009 |
+
"grad_norm": 0.361328125,
|
| 1010 |
+
"learning_rate": 5.549789490094304e-06,
|
| 1011 |
+
"loss": 1.0166,
|
| 1012 |
+
"step": 660
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 9.10958904109589,
|
| 1016 |
+
"grad_norm": 0.365234375,
|
| 1017 |
+
"learning_rate": 4.79144695714504e-06,
|
| 1018 |
+
"loss": 1.0251,
|
| 1019 |
+
"step": 665
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 9.178082191780822,
|
| 1023 |
+
"grad_norm": 0.390625,
|
| 1024 |
+
"learning_rate": 4.087525145291204e-06,
|
| 1025 |
+
"loss": 1.0213,
|
| 1026 |
+
"step": 670
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 9.246575342465754,
|
| 1030 |
+
"grad_norm": 0.404296875,
|
| 1031 |
+
"learning_rate": 3.4384264126337328e-06,
|
| 1032 |
+
"loss": 1.0181,
|
| 1033 |
+
"step": 675
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 9.315068493150685,
|
| 1037 |
+
"grad_norm": 0.357421875,
|
| 1038 |
+
"learning_rate": 2.8445217806824077e-06,
|
| 1039 |
+
"loss": 1.0202,
|
| 1040 |
+
"step": 680
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 9.383561643835616,
|
| 1044 |
+
"grad_norm": 0.361328125,
|
| 1045 |
+
"learning_rate": 2.30615072228183e-06,
|
| 1046 |
+
"loss": 1.0139,
|
| 1047 |
+
"step": 685
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 9.452054794520548,
|
| 1051 |
+
"grad_norm": 0.359375,
|
| 1052 |
+
"learning_rate": 1.8236209675705274e-06,
|
| 1053 |
+
"loss": 1.0292,
|
| 1054 |
+
"step": 690
|
| 1055 |
+
},
|
| 1056 |
+
{
|
| 1057 |
+
"epoch": 9.520547945205479,
|
| 1058 |
+
"grad_norm": 0.36328125,
|
| 1059 |
+
"learning_rate": 1.397208328083921e-06,
|
| 1060 |
+
"loss": 1.016,
|
| 1061 |
+
"step": 695
|
| 1062 |
+
},
|
| 1063 |
+
{
|
| 1064 |
+
"epoch": 9.58904109589041,
|
| 1065 |
+
"grad_norm": 0.35546875,
|
| 1066 |
+
"learning_rate": 1.0271565391018922e-06,
|
| 1067 |
+
"loss": 1.0206,
|
| 1068 |
+
"step": 700
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 9.657534246575342,
|
| 1072 |
+
"grad_norm": 0.369140625,
|
| 1073 |
+
"learning_rate": 7.136771203310245e-07,
|
| 1074 |
+
"loss": 1.0181,
|
| 1075 |
+
"step": 705
|
| 1076 |
+
},
|
| 1077 |
+
{
|
| 1078 |
+
"epoch": 9.726027397260275,
|
| 1079 |
+
"grad_norm": 0.359375,
|
| 1080 |
+
"learning_rate": 4.569492550008603e-07,
|
| 1081 |
+
"loss": 1.0123,
|
| 1082 |
+
"step": 710
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 9.794520547945206,
|
| 1086 |
+
"grad_norm": 0.3515625,
|
| 1087 |
+
"learning_rate": 2.5711968744382974e-07,
|
| 1088 |
+
"loss": 1.0101,
|
| 1089 |
+
"step": 715
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 9.863013698630137,
|
| 1093 |
+
"grad_norm": 0.3671875,
|
| 1094 |
+
"learning_rate": 1.143026392168789e-07,
|
| 1095 |
+
"loss": 1.0103,
|
| 1096 |
+
"step": 720
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 9.931506849315069,
|
| 1100 |
+
"grad_norm": 0.365234375,
|
| 1101 |
+
"learning_rate": 2.8579743813006432e-08,
|
| 1102 |
+
"loss": 1.0282,
|
| 1103 |
+
"step": 725
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 10.0,
|
| 1107 |
+
"grad_norm": 0.36328125,
|
| 1108 |
+
"learning_rate": 0.0,
|
| 1109 |
+
"loss": 1.0179,
|
| 1110 |
+
"step": 730
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 10.0,
|
| 1114 |
+
"eval_loss": 2.5617620944976807,
|
| 1115 |
+
"eval_runtime": 0.5453,
|
| 1116 |
+
"eval_samples_per_second": 18.339,
|
| 1117 |
+
"eval_steps_per_second": 1.834,
|
| 1118 |
+
"step": 730
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 10.0,
|
| 1122 |
+
"step": 730,
|
| 1123 |
+
"total_flos": 4.287825372721971e+17,
|
| 1124 |
+
"train_loss": 1.1969801510850044,
|
| 1125 |
+
"train_runtime": 3891.9077,
|
| 1126 |
+
"train_samples_per_second": 9.003,
|
| 1127 |
+
"train_steps_per_second": 0.188
|
| 1128 |
+
}
|
| 1129 |
+
],
|
| 1130 |
+
"logging_steps": 5,
|
| 1131 |
+
"max_steps": 730,
|
| 1132 |
+
"num_input_tokens_seen": 0,
|
| 1133 |
+
"num_train_epochs": 10,
|
| 1134 |
+
"save_steps": 100,
|
| 1135 |
+
"stateful_callbacks": {
|
| 1136 |
+
"TrainerControl": {
|
| 1137 |
+
"args": {
|
| 1138 |
+
"should_epoch_stop": false,
|
| 1139 |
+
"should_evaluate": false,
|
| 1140 |
+
"should_log": false,
|
| 1141 |
+
"should_save": true,
|
| 1142 |
+
"should_training_stop": false
|
| 1143 |
+
},
|
| 1144 |
+
"attributes": {}
|
| 1145 |
+
}
|
| 1146 |
+
},
|
| 1147 |
+
"total_flos": 4.287825372721971e+17,
|
| 1148 |
+
"train_batch_size": 8,
|
| 1149 |
+
"trial_name": null,
|
| 1150 |
+
"trial_params": null
|
| 1151 |
+
}
|