Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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: coding_llamaduo_result2
|
| 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 |
+
# coding_llamaduo_result2
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.2247
|
| 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: 2
|
| 44 |
+
- eval_batch_size: 2
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 2
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 8
|
| 50 |
+
- total_eval_batch_size: 4
|
| 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: 5
|
| 55 |
+
|
| 56 |
+
### Training results
|
| 57 |
+
|
| 58 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 59 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 60 |
+
| 0.8192 | 1.0 | 122 | 1.2006 |
|
| 61 |
+
| 0.6377 | 2.0 | 245 | 1.1304 |
|
| 62 |
+
| 0.5334 | 3.0 | 367 | 1.1456 |
|
| 63 |
+
| 0.4454 | 4.0 | 490 | 1.1935 |
|
| 64 |
+
| 0.408 | 4.98 | 610 | 1.2247 |
|
| 65 |
+
|
| 66 |
+
|
| 67 |
+
### Framework versions
|
| 68 |
+
|
| 69 |
+
- PEFT 0.7.1
|
| 70 |
+
- Transformers 4.39.3
|
| 71 |
+
- Pytorch 2.2.1+cu121
|
| 72 |
+
- Datasets 2.18.0
|
| 73 |
+
- Tokenizers 0.15.2
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 100060536
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6b808cfc275e2eb08c2745c6971280d3a476e2080eb98b2bd6dbd68953c2dd0a
|
| 3 |
size 100060536
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.98,
|
| 3 |
+
"train_loss": 1.826815680011374,
|
| 4 |
+
"train_runtime": 4235.156,
|
| 5 |
+
"train_samples": 13844,
|
| 6 |
+
"train_samples_per_second": 1.156,
|
| 7 |
+
"train_steps_per_second": 0.144
|
| 8 |
+
}
|
runs/Apr21_09-52-59_user-WS-C621E-SAGE-Series/events.out.tfevents.1713660853.user-WS-C621E-SAGE-Series.68748.0
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:db9760a15635bfed802eebf986ce65f8abeec5680508479b89100ee5f60b3f3c
|
| 3 |
+
size 32904
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.98,
|
| 3 |
+
"train_loss": 1.826815680011374,
|
| 4 |
+
"train_runtime": 4235.156,
|
| 5 |
+
"train_samples": 13844,
|
| 6 |
+
"train_samples_per_second": 1.156,
|
| 7 |
+
"train_steps_per_second": 0.144
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,931 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.979591836734694,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 610,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"grad_norm": 312.0,
|
| 14 |
+
"learning_rate": 3.278688524590164e-06,
|
| 15 |
+
"loss": 33.9196,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04,
|
| 20 |
+
"grad_norm": 354.0,
|
| 21 |
+
"learning_rate": 1.6393442622950818e-05,
|
| 22 |
+
"loss": 40.6749,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.08,
|
| 27 |
+
"grad_norm": 125.0,
|
| 28 |
+
"learning_rate": 3.2786885245901635e-05,
|
| 29 |
+
"loss": 27.6429,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.12,
|
| 34 |
+
"grad_norm": 29.875,
|
| 35 |
+
"learning_rate": 4.918032786885246e-05,
|
| 36 |
+
"loss": 19.7282,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16,
|
| 41 |
+
"grad_norm": 10.1875,
|
| 42 |
+
"learning_rate": 6.557377049180327e-05,
|
| 43 |
+
"loss": 16.6712,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.2,
|
| 48 |
+
"grad_norm": 5.53125,
|
| 49 |
+
"learning_rate": 8.19672131147541e-05,
|
| 50 |
+
"loss": 14.7944,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.24,
|
| 55 |
+
"grad_norm": 7.09375,
|
| 56 |
+
"learning_rate": 9.836065573770493e-05,
|
| 57 |
+
"loss": 14.0597,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.29,
|
| 62 |
+
"grad_norm": 11.875,
|
| 63 |
+
"learning_rate": 0.00011475409836065574,
|
| 64 |
+
"loss": 12.5767,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.33,
|
| 69 |
+
"grad_norm": 55.75,
|
| 70 |
+
"learning_rate": 0.00013114754098360654,
|
| 71 |
+
"loss": 9.9687,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.37,
|
| 76 |
+
"grad_norm": 9.5,
|
| 77 |
+
"learning_rate": 0.00014754098360655738,
|
| 78 |
+
"loss": 2.8926,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.41,
|
| 83 |
+
"grad_norm": 2.78125,
|
| 84 |
+
"learning_rate": 0.0001639344262295082,
|
| 85 |
+
"loss": 1.5807,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.45,
|
| 90 |
+
"grad_norm": 2.609375,
|
| 91 |
+
"learning_rate": 0.00018032786885245904,
|
| 92 |
+
"loss": 1.3009,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.49,
|
| 97 |
+
"grad_norm": 2.109375,
|
| 98 |
+
"learning_rate": 0.00019672131147540985,
|
| 99 |
+
"loss": 1.1905,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.53,
|
| 104 |
+
"grad_norm": 1.4921875,
|
| 105 |
+
"learning_rate": 0.00019997380453683512,
|
| 106 |
+
"loss": 1.1472,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.57,
|
| 111 |
+
"grad_norm": 3.5,
|
| 112 |
+
"learning_rate": 0.00019986740898848306,
|
| 113 |
+
"loss": 1.0454,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.61,
|
| 118 |
+
"grad_norm": 1.84375,
|
| 119 |
+
"learning_rate": 0.00019967926316561134,
|
| 120 |
+
"loss": 0.9724,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.65,
|
| 125 |
+
"grad_norm": 2.421875,
|
| 126 |
+
"learning_rate": 0.00019940952108215526,
|
| 127 |
+
"loss": 0.9521,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.69,
|
| 132 |
+
"grad_norm": 6.9375,
|
| 133 |
+
"learning_rate": 0.00019905840354577972,
|
| 134 |
+
"loss": 0.9019,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.73,
|
| 139 |
+
"grad_norm": 3.78125,
|
| 140 |
+
"learning_rate": 0.00019862619797712846,
|
| 141 |
+
"loss": 0.8665,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.78,
|
| 146 |
+
"grad_norm": 2.34375,
|
| 147 |
+
"learning_rate": 0.00019811325817454517,
|
| 148 |
+
"loss": 0.8629,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.82,
|
| 153 |
+
"grad_norm": 1.046875,
|
| 154 |
+
"learning_rate": 0.00019752000402445825,
|
| 155 |
+
"loss": 0.7888,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.86,
|
| 160 |
+
"grad_norm": 1.1875,
|
| 161 |
+
"learning_rate": 0.0001968469211576668,
|
| 162 |
+
"loss": 0.8141,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.9,
|
| 167 |
+
"grad_norm": 3.65625,
|
| 168 |
+
"learning_rate": 0.00019609456055180897,
|
| 169 |
+
"loss": 0.8212,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.94,
|
| 174 |
+
"grad_norm": 6.96875,
|
| 175 |
+
"learning_rate": 0.00019526353808033825,
|
| 176 |
+
"loss": 0.8073,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.98,
|
| 181 |
+
"grad_norm": 2.171875,
|
| 182 |
+
"learning_rate": 0.00019435453400837684,
|
| 183 |
+
"loss": 0.8192,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.0,
|
| 188 |
+
"eval_loss": 1.2006253004074097,
|
| 189 |
+
"eval_runtime": 1.1043,
|
| 190 |
+
"eval_samples_per_second": 1.811,
|
| 191 |
+
"eval_steps_per_second": 0.906,
|
| 192 |
+
"step": 122
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.02,
|
| 196 |
+
"grad_norm": 0.62890625,
|
| 197 |
+
"learning_rate": 0.00019336829243585856,
|
| 198 |
+
"loss": 0.7633,
|
| 199 |
+
"step": 125
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.06,
|
| 203 |
+
"grad_norm": 1.3125,
|
| 204 |
+
"learning_rate": 0.0001923056206884176,
|
| 205 |
+
"loss": 0.7091,
|
| 206 |
+
"step": 130
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.1,
|
| 210 |
+
"grad_norm": 0.578125,
|
| 211 |
+
"learning_rate": 0.00019116738865652134,
|
| 212 |
+
"loss": 0.6773,
|
| 213 |
+
"step": 135
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.14,
|
| 217 |
+
"grad_norm": 1.2578125,
|
| 218 |
+
"learning_rate": 0.00018995452808338823,
|
| 219 |
+
"loss": 0.7058,
|
| 220 |
+
"step": 140
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.18,
|
| 224 |
+
"grad_norm": 1.140625,
|
| 225 |
+
"learning_rate": 0.00018866803180227402,
|
| 226 |
+
"loss": 0.7189,
|
| 227 |
+
"step": 145
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.22,
|
| 231 |
+
"grad_norm": 1.71875,
|
| 232 |
+
"learning_rate": 0.00018730895292375018,
|
| 233 |
+
"loss": 0.6698,
|
| 234 |
+
"step": 150
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.27,
|
| 238 |
+
"grad_norm": 1.515625,
|
| 239 |
+
"learning_rate": 0.00018587840397364007,
|
| 240 |
+
"loss": 0.6941,
|
| 241 |
+
"step": 155
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.31,
|
| 245 |
+
"grad_norm": 0.482421875,
|
| 246 |
+
"learning_rate": 0.00018437755598231856,
|
| 247 |
+
"loss": 0.6511,
|
| 248 |
+
"step": 160
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.35,
|
| 252 |
+
"grad_norm": 1.3203125,
|
| 253 |
+
"learning_rate": 0.00018280763752612052,
|
| 254 |
+
"loss": 0.6672,
|
| 255 |
+
"step": 165
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.39,
|
| 259 |
+
"grad_norm": 0.72265625,
|
| 260 |
+
"learning_rate": 0.00018116993372164263,
|
| 261 |
+
"loss": 0.698,
|
| 262 |
+
"step": 170
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.43,
|
| 266 |
+
"grad_norm": 0.7890625,
|
| 267 |
+
"learning_rate": 0.0001794657851737625,
|
| 268 |
+
"loss": 0.6736,
|
| 269 |
+
"step": 175
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.47,
|
| 273 |
+
"grad_norm": 0.5234375,
|
| 274 |
+
"learning_rate": 0.00017769658687823525,
|
| 275 |
+
"loss": 0.651,
|
| 276 |
+
"step": 180
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.51,
|
| 280 |
+
"grad_norm": 0.57421875,
|
| 281 |
+
"learning_rate": 0.0001758637870797665,
|
| 282 |
+
"loss": 0.6809,
|
| 283 |
+
"step": 185
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.55,
|
| 287 |
+
"grad_norm": 1.1875,
|
| 288 |
+
"learning_rate": 0.0001739688860864967,
|
| 289 |
+
"loss": 0.6331,
|
| 290 |
+
"step": 190
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.59,
|
| 294 |
+
"grad_norm": 0.8359375,
|
| 295 |
+
"learning_rate": 0.00017201343504186643,
|
| 296 |
+
"loss": 0.6562,
|
| 297 |
+
"step": 195
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.63,
|
| 301 |
+
"grad_norm": 1.5625,
|
| 302 |
+
"learning_rate": 0.00016999903465486914,
|
| 303 |
+
"loss": 0.6416,
|
| 304 |
+
"step": 200
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.67,
|
| 308 |
+
"grad_norm": 1.3359375,
|
| 309 |
+
"learning_rate": 0.00016792733388972932,
|
| 310 |
+
"loss": 0.6957,
|
| 311 |
+
"step": 205
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.71,
|
| 315 |
+
"grad_norm": 0.93359375,
|
| 316 |
+
"learning_rate": 0.0001658000286160805,
|
| 317 |
+
"loss": 0.6857,
|
| 318 |
+
"step": 210
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.76,
|
| 322 |
+
"grad_norm": 1.0703125,
|
| 323 |
+
"learning_rate": 0.00016361886022074612,
|
| 324 |
+
"loss": 0.6589,
|
| 325 |
+
"step": 215
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.8,
|
| 329 |
+
"grad_norm": 0.62109375,
|
| 330 |
+
"learning_rate": 0.0001613856141822612,
|
| 331 |
+
"loss": 0.6635,
|
| 332 |
+
"step": 220
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.84,
|
| 336 |
+
"grad_norm": 0.38671875,
|
| 337 |
+
"learning_rate": 0.00015910211860930065,
|
| 338 |
+
"loss": 0.6252,
|
| 339 |
+
"step": 225
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.88,
|
| 343 |
+
"grad_norm": 0.78515625,
|
| 344 |
+
"learning_rate": 0.0001567702427442113,
|
| 345 |
+
"loss": 0.6187,
|
| 346 |
+
"step": 230
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.92,
|
| 350 |
+
"grad_norm": 1.8984375,
|
| 351 |
+
"learning_rate": 0.00015439189543287247,
|
| 352 |
+
"loss": 0.6194,
|
| 353 |
+
"step": 235
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.96,
|
| 357 |
+
"grad_norm": 1.328125,
|
| 358 |
+
"learning_rate": 0.00015196902356213714,
|
| 359 |
+
"loss": 0.6396,
|
| 360 |
+
"step": 240
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 2.0,
|
| 364 |
+
"grad_norm": 0.87109375,
|
| 365 |
+
"learning_rate": 0.0001495036104661337,
|
| 366 |
+
"loss": 0.6377,
|
| 367 |
+
"step": 245
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 2.0,
|
| 371 |
+
"eval_loss": 1.1303967237472534,
|
| 372 |
+
"eval_runtime": 1.0903,
|
| 373 |
+
"eval_samples_per_second": 1.834,
|
| 374 |
+
"eval_steps_per_second": 0.917,
|
| 375 |
+
"step": 245
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.04,
|
| 379 |
+
"grad_norm": 1.0234375,
|
| 380 |
+
"learning_rate": 0.000146997674302732,
|
| 381 |
+
"loss": 0.5446,
|
| 382 |
+
"step": 250
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.08,
|
| 386 |
+
"grad_norm": 0.76171875,
|
| 387 |
+
"learning_rate": 0.0001444532664015037,
|
| 388 |
+
"loss": 0.5915,
|
| 389 |
+
"step": 255
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.12,
|
| 393 |
+
"grad_norm": 1.28125,
|
| 394 |
+
"learning_rate": 0.0001418724695845277,
|
| 395 |
+
"loss": 0.5521,
|
| 396 |
+
"step": 260
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.16,
|
| 400 |
+
"grad_norm": 1.3203125,
|
| 401 |
+
"learning_rate": 0.0001392573964614172,
|
| 402 |
+
"loss": 0.5279,
|
| 403 |
+
"step": 265
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.2,
|
| 407 |
+
"grad_norm": 0.470703125,
|
| 408 |
+
"learning_rate": 0.00013661018769996227,
|
| 409 |
+
"loss": 0.558,
|
| 410 |
+
"step": 270
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.24,
|
| 414 |
+
"grad_norm": 0.82421875,
|
| 415 |
+
"learning_rate": 0.00013393301027380476,
|
| 416 |
+
"loss": 0.5422,
|
| 417 |
+
"step": 275
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.29,
|
| 421 |
+
"grad_norm": 0.53125,
|
| 422 |
+
"learning_rate": 0.00013122805568857948,
|
| 423 |
+
"loss": 0.5401,
|
| 424 |
+
"step": 280
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.33,
|
| 428 |
+
"grad_norm": 0.6171875,
|
| 429 |
+
"learning_rate": 0.00012849753818797353,
|
| 430 |
+
"loss": 0.5458,
|
| 431 |
+
"step": 285
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.37,
|
| 435 |
+
"grad_norm": 0.47265625,
|
| 436 |
+
"learning_rate": 0.00012574369294117297,
|
| 437 |
+
"loss": 0.5309,
|
| 438 |
+
"step": 290
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.41,
|
| 442 |
+
"grad_norm": 0.490234375,
|
| 443 |
+
"learning_rate": 0.0001229687742131796,
|
| 444 |
+
"loss": 0.5276,
|
| 445 |
+
"step": 295
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.45,
|
| 449 |
+
"grad_norm": 0.6796875,
|
| 450 |
+
"learning_rate": 0.00012017505351949661,
|
| 451 |
+
"loss": 0.5322,
|
| 452 |
+
"step": 300
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.49,
|
| 456 |
+
"grad_norm": 0.53515625,
|
| 457 |
+
"learning_rate": 0.00011736481776669306,
|
| 458 |
+
"loss": 0.5242,
|
| 459 |
+
"step": 305
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.53,
|
| 463 |
+
"grad_norm": 0.55859375,
|
| 464 |
+
"learning_rate": 0.00011454036738036899,
|
| 465 |
+
"loss": 0.5104,
|
| 466 |
+
"step": 310
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.57,
|
| 470 |
+
"grad_norm": 1.0234375,
|
| 471 |
+
"learning_rate": 0.00011170401442205474,
|
| 472 |
+
"loss": 0.6306,
|
| 473 |
+
"step": 315
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.61,
|
| 477 |
+
"grad_norm": 0.5078125,
|
| 478 |
+
"learning_rate": 0.00010885808069658451,
|
| 479 |
+
"loss": 0.5246,
|
| 480 |
+
"step": 320
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.65,
|
| 484 |
+
"grad_norm": 0.86328125,
|
| 485 |
+
"learning_rate": 0.00010600489585149484,
|
| 486 |
+
"loss": 0.5307,
|
| 487 |
+
"step": 325
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.69,
|
| 491 |
+
"grad_norm": 0.546875,
|
| 492 |
+
"learning_rate": 0.0001031467954700025,
|
| 493 |
+
"loss": 0.523,
|
| 494 |
+
"step": 330
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.73,
|
| 498 |
+
"grad_norm": 0.44921875,
|
| 499 |
+
"learning_rate": 0.00010028611915912405,
|
| 500 |
+
"loss": 0.5449,
|
| 501 |
+
"step": 335
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.78,
|
| 505 |
+
"grad_norm": 0.443359375,
|
| 506 |
+
"learning_rate": 9.742520863450115e-05,
|
| 507 |
+
"loss": 0.5226,
|
| 508 |
+
"step": 340
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.82,
|
| 512 |
+
"grad_norm": 0.466796875,
|
| 513 |
+
"learning_rate": 9.456640580350018e-05,
|
| 514 |
+
"loss": 0.5173,
|
| 515 |
+
"step": 345
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.86,
|
| 519 |
+
"grad_norm": 0.76953125,
|
| 520 |
+
"learning_rate": 9.171205084815444e-05,
|
| 521 |
+
"loss": 0.5281,
|
| 522 |
+
"step": 350
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.9,
|
| 526 |
+
"grad_norm": 0.6796875,
|
| 527 |
+
"learning_rate": 8.886448030951912e-05,
|
| 528 |
+
"loss": 0.543,
|
| 529 |
+
"step": 355
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.94,
|
| 533 |
+
"grad_norm": 0.427734375,
|
| 534 |
+
"learning_rate": 8.602602517500651e-05,
|
| 535 |
+
"loss": 0.5459,
|
| 536 |
+
"step": 360
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.98,
|
| 540 |
+
"grad_norm": 0.5,
|
| 541 |
+
"learning_rate": 8.319900897026733e-05,
|
| 542 |
+
"loss": 0.5334,
|
| 543 |
+
"step": 365
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 3.0,
|
| 547 |
+
"eval_loss": 1.1456480026245117,
|
| 548 |
+
"eval_runtime": 1.0913,
|
| 549 |
+
"eval_samples_per_second": 1.833,
|
| 550 |
+
"eval_steps_per_second": 0.916,
|
| 551 |
+
"step": 367
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 3.02,
|
| 555 |
+
"grad_norm": 0.494140625,
|
| 556 |
+
"learning_rate": 8.038574585718032e-05,
|
| 557 |
+
"loss": 0.5025,
|
| 558 |
+
"step": 370
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 3.06,
|
| 562 |
+
"grad_norm": 0.69921875,
|
| 563 |
+
"learning_rate": 7.758853873950676e-05,
|
| 564 |
+
"loss": 0.4247,
|
| 565 |
+
"step": 375
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 3.1,
|
| 569 |
+
"grad_norm": 0.48828125,
|
| 570 |
+
"learning_rate": 7.480967737776089e-05,
|
| 571 |
+
"loss": 0.4551,
|
| 572 |
+
"step": 380
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 3.14,
|
| 576 |
+
"grad_norm": 0.58984375,
|
| 577 |
+
"learning_rate": 7.205143651483906e-05,
|
| 578 |
+
"loss": 0.4667,
|
| 579 |
+
"step": 385
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.18,
|
| 583 |
+
"grad_norm": 0.59375,
|
| 584 |
+
"learning_rate": 6.931607401394229e-05,
|
| 585 |
+
"loss": 0.4581,
|
| 586 |
+
"step": 390
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 3.22,
|
| 590 |
+
"grad_norm": 0.7421875,
|
| 591 |
+
"learning_rate": 6.660582901031621e-05,
|
| 592 |
+
"loss": 0.4441,
|
| 593 |
+
"step": 395
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 3.27,
|
| 597 |
+
"grad_norm": 0.546875,
|
| 598 |
+
"learning_rate": 6.392292007832168e-05,
|
| 599 |
+
"loss": 0.4247,
|
| 600 |
+
"step": 400
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 3.31,
|
| 604 |
+
"grad_norm": 0.66796875,
|
| 605 |
+
"learning_rate": 6.126954341533599e-05,
|
| 606 |
+
"loss": 0.4214,
|
| 607 |
+
"step": 405
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 3.35,
|
| 611 |
+
"grad_norm": 0.48046875,
|
| 612 |
+
"learning_rate": 5.864787104397194e-05,
|
| 613 |
+
"loss": 0.4413,
|
| 614 |
+
"step": 410
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.39,
|
| 618 |
+
"grad_norm": 0.65234375,
|
| 619 |
+
"learning_rate": 5.6060049034085815e-05,
|
| 620 |
+
"loss": 0.4648,
|
| 621 |
+
"step": 415
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 3.43,
|
| 625 |
+
"grad_norm": 0.53125,
|
| 626 |
+
"learning_rate": 5.3508195746030296e-05,
|
| 627 |
+
"loss": 0.5039,
|
| 628 |
+
"step": 420
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 3.47,
|
| 632 |
+
"grad_norm": 0.498046875,
|
| 633 |
+
"learning_rate": 5.099440009658957e-05,
|
| 634 |
+
"loss": 0.4493,
|
| 635 |
+
"step": 425
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 3.51,
|
| 639 |
+
"grad_norm": 0.43359375,
|
| 640 |
+
"learning_rate": 4.852071984901696e-05,
|
| 641 |
+
"loss": 0.4343,
|
| 642 |
+
"step": 430
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 3.55,
|
| 646 |
+
"grad_norm": 0.52734375,
|
| 647 |
+
"learning_rate": 4.6089179928574487e-05,
|
| 648 |
+
"loss": 0.4354,
|
| 649 |
+
"step": 435
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 3.59,
|
| 653 |
+
"grad_norm": 0.48046875,
|
| 654 |
+
"learning_rate": 4.3701770764953124e-05,
|
| 655 |
+
"loss": 0.4694,
|
| 656 |
+
"step": 440
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 3.63,
|
| 660 |
+
"grad_norm": 0.431640625,
|
| 661 |
+
"learning_rate": 4.136044666293044e-05,
|
| 662 |
+
"loss": 0.4446,
|
| 663 |
+
"step": 445
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 3.67,
|
| 667 |
+
"grad_norm": 0.447265625,
|
| 668 |
+
"learning_rate": 3.9067124202600194e-05,
|
| 669 |
+
"loss": 0.4398,
|
| 670 |
+
"step": 450
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 3.71,
|
| 674 |
+
"grad_norm": 0.4609375,
|
| 675 |
+
"learning_rate": 3.6823680670482485e-05,
|
| 676 |
+
"loss": 0.4449,
|
| 677 |
+
"step": 455
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 3.76,
|
| 681 |
+
"grad_norm": 0.4765625,
|
| 682 |
+
"learning_rate": 3.463195252279939e-05,
|
| 683 |
+
"loss": 0.4443,
|
| 684 |
+
"step": 460
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 3.8,
|
| 688 |
+
"grad_norm": 0.423828125,
|
| 689 |
+
"learning_rate": 3.24937338821738e-05,
|
| 690 |
+
"loss": 0.4346,
|
| 691 |
+
"step": 465
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 3.84,
|
| 695 |
+
"grad_norm": 0.4453125,
|
| 696 |
+
"learning_rate": 3.0410775068981613e-05,
|
| 697 |
+
"loss": 0.4423,
|
| 698 |
+
"step": 470
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 3.88,
|
| 702 |
+
"grad_norm": 0.828125,
|
| 703 |
+
"learning_rate": 2.8384781168560693e-05,
|
| 704 |
+
"loss": 0.4304,
|
| 705 |
+
"step": 475
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 3.92,
|
| 709 |
+
"grad_norm": 0.490234375,
|
| 710 |
+
"learning_rate": 2.6417410635448013e-05,
|
| 711 |
+
"loss": 0.443,
|
| 712 |
+
"step": 480
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 3.96,
|
| 716 |
+
"grad_norm": 0.46875,
|
| 717 |
+
"learning_rate": 2.451027393578863e-05,
|
| 718 |
+
"loss": 0.4392,
|
| 719 |
+
"step": 485
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 4.0,
|
| 723 |
+
"grad_norm": 0.5,
|
| 724 |
+
"learning_rate": 2.2664932229027024e-05,
|
| 725 |
+
"loss": 0.4454,
|
| 726 |
+
"step": 490
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 4.0,
|
| 730 |
+
"eval_loss": 1.1934518814086914,
|
| 731 |
+
"eval_runtime": 1.0901,
|
| 732 |
+
"eval_samples_per_second": 1.835,
|
| 733 |
+
"eval_steps_per_second": 0.917,
|
| 734 |
+
"step": 490
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 4.04,
|
| 738 |
+
"grad_norm": 0.45703125,
|
| 739 |
+
"learning_rate": 2.088289608996071e-05,
|
| 740 |
+
"loss": 0.3925,
|
| 741 |
+
"step": 495
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 4.08,
|
| 745 |
+
"grad_norm": 0.4921875,
|
| 746 |
+
"learning_rate": 1.9165624272201567e-05,
|
| 747 |
+
"loss": 0.403,
|
| 748 |
+
"step": 500
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 4.12,
|
| 752 |
+
"grad_norm": 0.5,
|
| 753 |
+
"learning_rate": 1.7514522514057553e-05,
|
| 754 |
+
"loss": 0.3824,
|
| 755 |
+
"step": 505
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 4.16,
|
| 759 |
+
"grad_norm": 0.462890625,
|
| 760 |
+
"learning_rate": 1.593094238781191e-05,
|
| 761 |
+
"loss": 0.3818,
|
| 762 |
+
"step": 510
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 4.2,
|
| 766 |
+
"grad_norm": 0.47265625,
|
| 767 |
+
"learning_rate": 1.4416180193342144e-05,
|
| 768 |
+
"loss": 0.4017,
|
| 769 |
+
"step": 515
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 4.24,
|
| 773 |
+
"grad_norm": 0.4609375,
|
| 774 |
+
"learning_rate": 1.2971475896984475e-05,
|
| 775 |
+
"loss": 0.4047,
|
| 776 |
+
"step": 520
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 4.29,
|
| 780 |
+
"grad_norm": 0.466796875,
|
| 781 |
+
"learning_rate": 1.1598012116512092e-05,
|
| 782 |
+
"loss": 0.3735,
|
| 783 |
+
"step": 525
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 4.33,
|
| 787 |
+
"grad_norm": 0.47265625,
|
| 788 |
+
"learning_rate": 1.0296913153058163e-05,
|
| 789 |
+
"loss": 0.3954,
|
| 790 |
+
"step": 530
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 4.37,
|
| 794 |
+
"grad_norm": 0.49609375,
|
| 795 |
+
"learning_rate": 9.069244070776428e-06,
|
| 796 |
+
"loss": 0.3955,
|
| 797 |
+
"step": 535
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 4.41,
|
| 801 |
+
"grad_norm": 0.5,
|
| 802 |
+
"learning_rate": 7.916009824992254e-06,
|
| 803 |
+
"loss": 0.3824,
|
| 804 |
+
"step": 540
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 4.45,
|
| 808 |
+
"grad_norm": 0.53125,
|
| 809 |
+
"learning_rate": 6.838154439558253e-06,
|
| 810 |
+
"loss": 0.3846,
|
| 811 |
+
"step": 545
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 4.49,
|
| 815 |
+
"grad_norm": 0.455078125,
|
| 816 |
+
"learning_rate": 5.836560234087418e-06,
|
| 817 |
+
"loss": 0.4087,
|
| 818 |
+
"step": 550
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 4.53,
|
| 822 |
+
"grad_norm": 0.455078125,
|
| 823 |
+
"learning_rate": 4.912047101696848e-06,
|
| 824 |
+
"loss": 0.3785,
|
| 825 |
+
"step": 555
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 4.57,
|
| 829 |
+
"grad_norm": 0.48828125,
|
| 830 |
+
"learning_rate": 4.065371837853016e-06,
|
| 831 |
+
"loss": 0.4013,
|
| 832 |
+
"step": 560
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 4.61,
|
| 836 |
+
"grad_norm": 0.55859375,
|
| 837 |
+
"learning_rate": 3.2972275208679625e-06,
|
| 838 |
+
"loss": 0.4005,
|
| 839 |
+
"step": 565
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 4.65,
|
| 843 |
+
"grad_norm": 0.498046875,
|
| 844 |
+
"learning_rate": 2.608242944553607e-06,
|
| 845 |
+
"loss": 0.3842,
|
| 846 |
+
"step": 570
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 4.69,
|
| 850 |
+
"grad_norm": 0.470703125,
|
| 851 |
+
"learning_rate": 1.9989821034986032e-06,
|
| 852 |
+
"loss": 0.3791,
|
| 853 |
+
"step": 575
|
| 854 |
+
},
|
| 855 |
+
{
|
| 856 |
+
"epoch": 4.73,
|
| 857 |
+
"grad_norm": 0.494140625,
|
| 858 |
+
"learning_rate": 1.4699437313891007e-06,
|
| 859 |
+
"loss": 0.398,
|
| 860 |
+
"step": 580
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 4.78,
|
| 864 |
+
"grad_norm": 0.482421875,
|
| 865 |
+
"learning_rate": 1.02156089275125e-06,
|
| 866 |
+
"loss": 0.3836,
|
| 867 |
+
"step": 585
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 4.82,
|
| 871 |
+
"grad_norm": 0.55078125,
|
| 872 |
+
"learning_rate": 6.542006284497304e-07,
|
| 873 |
+
"loss": 0.3827,
|
| 874 |
+
"step": 590
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 4.86,
|
| 878 |
+
"grad_norm": 0.4296875,
|
| 879 |
+
"learning_rate": 3.681636552324452e-07,
|
| 880 |
+
"loss": 0.3846,
|
| 881 |
+
"step": 595
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 4.9,
|
| 885 |
+
"grad_norm": 0.4609375,
|
| 886 |
+
"learning_rate": 1.6368411956741991e-07,
|
| 887 |
+
"loss": 0.4347,
|
| 888 |
+
"step": 600
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 4.94,
|
| 892 |
+
"grad_norm": 0.47265625,
|
| 893 |
+
"learning_rate": 4.092940597322237e-08,
|
| 894 |
+
"loss": 0.3899,
|
| 895 |
+
"step": 605
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 4.98,
|
| 899 |
+
"grad_norm": 0.458984375,
|
| 900 |
+
"learning_rate": 0.0,
|
| 901 |
+
"loss": 0.408,
|
| 902 |
+
"step": 610
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 4.98,
|
| 906 |
+
"eval_loss": 1.2246639728546143,
|
| 907 |
+
"eval_runtime": 1.1185,
|
| 908 |
+
"eval_samples_per_second": 1.788,
|
| 909 |
+
"eval_steps_per_second": 0.894,
|
| 910 |
+
"step": 610
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 4.98,
|
| 914 |
+
"step": 610,
|
| 915 |
+
"total_flos": 4.678055725535068e+17,
|
| 916 |
+
"train_loss": 1.826815680011374,
|
| 917 |
+
"train_runtime": 4235.156,
|
| 918 |
+
"train_samples_per_second": 1.156,
|
| 919 |
+
"train_steps_per_second": 0.144
|
| 920 |
+
}
|
| 921 |
+
],
|
| 922 |
+
"logging_steps": 5,
|
| 923 |
+
"max_steps": 610,
|
| 924 |
+
"num_input_tokens_seen": 0,
|
| 925 |
+
"num_train_epochs": 5,
|
| 926 |
+
"save_steps": 100,
|
| 927 |
+
"total_flos": 4.678055725535068e+17,
|
| 928 |
+
"train_batch_size": 2,
|
| 929 |
+
"trial_name": null,
|
| 930 |
+
"trial_params": null
|
| 931 |
+
}
|