Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
license: llama3.2
|
| 4 |
+
base_model: meta-llama/Llama-3.2-3B
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
datasets:
|
| 10 |
+
- generator
|
| 11 |
+
model-index:
|
| 12 |
+
- name: llama3-3b-closedqa-gpt4o-100k
|
| 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 |
+
# llama3-3b-closedqa-gpt4o-100k
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 2.3769
|
| 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: 16
|
| 44 |
+
- eval_batch_size: 16
|
| 45 |
+
- seed: 42
|
| 46 |
+
- distributed_type: multi-GPU
|
| 47 |
+
- num_devices: 8
|
| 48 |
+
- gradient_accumulation_steps: 2
|
| 49 |
+
- total_train_batch_size: 256
|
| 50 |
+
- total_eval_batch_size: 128
|
| 51 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 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 |
+
| 1.471 | 1.0 | 64 | 2.3854 |
|
| 61 |
+
| 1.362 | 2.0 | 128 | 2.3754 |
|
| 62 |
+
| 1.3229 | 3.0 | 192 | 2.3740 |
|
| 63 |
+
| 1.2996 | 4.0 | 256 | 2.3706 |
|
| 64 |
+
| 1.2878 | 5.0 | 320 | 2.3755 |
|
| 65 |
+
| 1.2746 | 6.0 | 384 | 2.3722 |
|
| 66 |
+
| 1.2617 | 7.0 | 448 | 2.3756 |
|
| 67 |
+
| 1.2497 | 8.0 | 512 | 2.3754 |
|
| 68 |
+
| 1.2549 | 9.0 | 576 | 2.3762 |
|
| 69 |
+
| 1.2494 | 10.0 | 640 | 2.3769 |
|
| 70 |
+
|
| 71 |
+
|
| 72 |
+
### Framework versions
|
| 73 |
+
|
| 74 |
+
- PEFT 0.15.1
|
| 75 |
+
- Transformers 4.50.3
|
| 76 |
+
- Pytorch 2.6.0+cu124
|
| 77 |
+
- Datasets 3.5.0
|
| 78 |
+
- Tokenizers 0.21.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1612749744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d846c3d7d8d294a4f449243d354efe697a0682a73620ffab23df4c1a459ec1f2
|
| 3 |
size 1612749744
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 2.846679860191953e+18,
|
| 4 |
+
"train_loss": 1.3252380434423685,
|
| 5 |
+
"train_runtime": 3360.483,
|
| 6 |
+
"train_samples": 111440,
|
| 7 |
+
"train_samples_per_second": 48.669,
|
| 8 |
+
"train_steps_per_second": 0.19
|
| 9 |
+
}
|
runs/Apr01_01-14-14_green-face-echoes-fin-01/events.out.tfevents.1743470256.green-face-echoes-fin-01.27822.0
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:bf2fa52dfa76cb05648f8b68caea459d3d5b0ac69a8096c16d589147c1391d48
|
| 3 |
+
size 36440
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 2.846679860191953e+18,
|
| 4 |
+
"train_loss": 1.3252380434423685,
|
| 5 |
+
"train_runtime": 3360.483,
|
| 6 |
+
"train_samples": 111440,
|
| 7 |
+
"train_samples_per_second": 48.669,
|
| 8 |
+
"train_steps_per_second": 0.19
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1026 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 640,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.015625,
|
| 14 |
+
"grad_norm": 0.5563601851463318,
|
| 15 |
+
"learning_rate": 3.125e-06,
|
| 16 |
+
"loss": 1.837,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.078125,
|
| 21 |
+
"grad_norm": 0.5579909682273865,
|
| 22 |
+
"learning_rate": 1.5625e-05,
|
| 23 |
+
"loss": 1.832,
|
| 24 |
+
"step": 5
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.15625,
|
| 28 |
+
"grad_norm": 0.3869417607784271,
|
| 29 |
+
"learning_rate": 3.125e-05,
|
| 30 |
+
"loss": 1.8211,
|
| 31 |
+
"step": 10
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.234375,
|
| 35 |
+
"grad_norm": 0.2597704827785492,
|
| 36 |
+
"learning_rate": 4.6875e-05,
|
| 37 |
+
"loss": 1.791,
|
| 38 |
+
"step": 15
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.3125,
|
| 42 |
+
"grad_norm": 0.2681010663509369,
|
| 43 |
+
"learning_rate": 6.25e-05,
|
| 44 |
+
"loss": 1.7632,
|
| 45 |
+
"step": 20
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.390625,
|
| 49 |
+
"grad_norm": 0.2458464354276657,
|
| 50 |
+
"learning_rate": 7.8125e-05,
|
| 51 |
+
"loss": 1.7142,
|
| 52 |
+
"step": 25
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.46875,
|
| 56 |
+
"grad_norm": 0.2320868819952011,
|
| 57 |
+
"learning_rate": 9.375e-05,
|
| 58 |
+
"loss": 1.662,
|
| 59 |
+
"step": 30
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.546875,
|
| 63 |
+
"grad_norm": 0.2217356562614441,
|
| 64 |
+
"learning_rate": 0.000109375,
|
| 65 |
+
"loss": 1.606,
|
| 66 |
+
"step": 35
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.625,
|
| 70 |
+
"grad_norm": 0.17081762850284576,
|
| 71 |
+
"learning_rate": 0.000125,
|
| 72 |
+
"loss": 1.5636,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.703125,
|
| 77 |
+
"grad_norm": 0.13257570564746857,
|
| 78 |
+
"learning_rate": 0.00014062500000000002,
|
| 79 |
+
"loss": 1.5291,
|
| 80 |
+
"step": 45
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.78125,
|
| 84 |
+
"grad_norm": 0.12237544357776642,
|
| 85 |
+
"learning_rate": 0.00015625,
|
| 86 |
+
"loss": 1.5163,
|
| 87 |
+
"step": 50
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.859375,
|
| 91 |
+
"grad_norm": 0.10626661777496338,
|
| 92 |
+
"learning_rate": 0.00017187500000000002,
|
| 93 |
+
"loss": 1.4857,
|
| 94 |
+
"step": 55
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.9375,
|
| 98 |
+
"grad_norm": 0.1030210331082344,
|
| 99 |
+
"learning_rate": 0.0001875,
|
| 100 |
+
"loss": 1.471,
|
| 101 |
+
"step": 60
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.0,
|
| 105 |
+
"eval_loss": 2.3853657245635986,
|
| 106 |
+
"eval_runtime": 0.8929,
|
| 107 |
+
"eval_samples_per_second": 6.72,
|
| 108 |
+
"eval_steps_per_second": 1.12,
|
| 109 |
+
"step": 64
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.015625,
|
| 113 |
+
"grad_norm": 0.1105547621846199,
|
| 114 |
+
"learning_rate": 0.00019999851261394218,
|
| 115 |
+
"loss": 1.4547,
|
| 116 |
+
"step": 65
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.09375,
|
| 120 |
+
"grad_norm": 0.1074226126074791,
|
| 121 |
+
"learning_rate": 0.00019994645874763658,
|
| 122 |
+
"loss": 1.4433,
|
| 123 |
+
"step": 70
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.171875,
|
| 127 |
+
"grad_norm": 0.1031486839056015,
|
| 128 |
+
"learning_rate": 0.00019982007981886847,
|
| 129 |
+
"loss": 1.4319,
|
| 130 |
+
"step": 75
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.25,
|
| 134 |
+
"grad_norm": 0.09431572258472443,
|
| 135 |
+
"learning_rate": 0.00019961946980917456,
|
| 136 |
+
"loss": 1.4231,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.328125,
|
| 141 |
+
"grad_norm": 0.10248947143554688,
|
| 142 |
+
"learning_rate": 0.00019934477790194445,
|
| 143 |
+
"loss": 1.414,
|
| 144 |
+
"step": 85
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.40625,
|
| 148 |
+
"grad_norm": 0.10952038317918777,
|
| 149 |
+
"learning_rate": 0.00019899620837148077,
|
| 150 |
+
"loss": 1.3994,
|
| 151 |
+
"step": 90
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.484375,
|
| 155 |
+
"grad_norm": 0.11935596913099289,
|
| 156 |
+
"learning_rate": 0.0001985740204310909,
|
| 157 |
+
"loss": 1.3933,
|
| 158 |
+
"step": 95
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.5625,
|
| 162 |
+
"grad_norm": 0.13777290284633636,
|
| 163 |
+
"learning_rate": 0.00019807852804032305,
|
| 164 |
+
"loss": 1.3891,
|
| 165 |
+
"step": 100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.640625,
|
| 169 |
+
"grad_norm": 0.10964534431695938,
|
| 170 |
+
"learning_rate": 0.00019751009967149087,
|
| 171 |
+
"loss": 1.3755,
|
| 172 |
+
"step": 105
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.71875,
|
| 176 |
+
"grad_norm": 0.10717500746250153,
|
| 177 |
+
"learning_rate": 0.00019686915803565934,
|
| 178 |
+
"loss": 1.3811,
|
| 179 |
+
"step": 110
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.796875,
|
| 183 |
+
"grad_norm": 0.11473763734102249,
|
| 184 |
+
"learning_rate": 0.0001961561797682962,
|
| 185 |
+
"loss": 1.3661,
|
| 186 |
+
"step": 115
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.875,
|
| 190 |
+
"grad_norm": 0.11382485926151276,
|
| 191 |
+
"learning_rate": 0.0001953716950748227,
|
| 192 |
+
"loss": 1.3596,
|
| 193 |
+
"step": 120
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 1.953125,
|
| 197 |
+
"grad_norm": 0.12709280848503113,
|
| 198 |
+
"learning_rate": 0.0001945162873363268,
|
| 199 |
+
"loss": 1.362,
|
| 200 |
+
"step": 125
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.0,
|
| 204 |
+
"eval_loss": 2.375438690185547,
|
| 205 |
+
"eval_runtime": 0.8905,
|
| 206 |
+
"eval_samples_per_second": 6.738,
|
| 207 |
+
"eval_steps_per_second": 1.123,
|
| 208 |
+
"step": 128
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.03125,
|
| 212 |
+
"grad_norm": 0.1248418539762497,
|
| 213 |
+
"learning_rate": 0.0001935905926757326,
|
| 214 |
+
"loss": 1.3543,
|
| 215 |
+
"step": 130
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.109375,
|
| 219 |
+
"grad_norm": 0.13678818941116333,
|
| 220 |
+
"learning_rate": 0.00019259529948474833,
|
| 221 |
+
"loss": 1.3531,
|
| 222 |
+
"step": 135
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.1875,
|
| 226 |
+
"grad_norm": 0.12561410665512085,
|
| 227 |
+
"learning_rate": 0.00019153114791194473,
|
| 228 |
+
"loss": 1.3382,
|
| 229 |
+
"step": 140
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.265625,
|
| 233 |
+
"grad_norm": 0.14160117506980896,
|
| 234 |
+
"learning_rate": 0.00019039892931234435,
|
| 235 |
+
"loss": 1.3388,
|
| 236 |
+
"step": 145
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.34375,
|
| 240 |
+
"grad_norm": 0.1499018520116806,
|
| 241 |
+
"learning_rate": 0.00018919948565893142,
|
| 242 |
+
"loss": 1.3392,
|
| 243 |
+
"step": 150
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.421875,
|
| 247 |
+
"grad_norm": 0.14188511669635773,
|
| 248 |
+
"learning_rate": 0.00018793370891651972,
|
| 249 |
+
"loss": 1.3406,
|
| 250 |
+
"step": 155
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.5,
|
| 254 |
+
"grad_norm": 0.12685342133045197,
|
| 255 |
+
"learning_rate": 0.00018660254037844388,
|
| 256 |
+
"loss": 1.3312,
|
| 257 |
+
"step": 160
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 2.578125,
|
| 261 |
+
"grad_norm": 0.1338503062725067,
|
| 262 |
+
"learning_rate": 0.00018520696996656788,
|
| 263 |
+
"loss": 1.3386,
|
| 264 |
+
"step": 165
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 2.65625,
|
| 268 |
+
"grad_norm": 0.1542833298444748,
|
| 269 |
+
"learning_rate": 0.0001837480354951308,
|
| 270 |
+
"loss": 1.3339,
|
| 271 |
+
"step": 170
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 2.734375,
|
| 275 |
+
"grad_norm": 0.1419745534658432,
|
| 276 |
+
"learning_rate": 0.00018222682189897752,
|
| 277 |
+
"loss": 1.3256,
|
| 278 |
+
"step": 175
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.8125,
|
| 282 |
+
"grad_norm": 0.12559981644153595,
|
| 283 |
+
"learning_rate": 0.00018064446042674828,
|
| 284 |
+
"loss": 1.3187,
|
| 285 |
+
"step": 180
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 2.890625,
|
| 289 |
+
"grad_norm": 0.1430642306804657,
|
| 290 |
+
"learning_rate": 0.0001790021277996269,
|
| 291 |
+
"loss": 1.3193,
|
| 292 |
+
"step": 185
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 2.96875,
|
| 296 |
+
"grad_norm": 0.12510241568088531,
|
| 297 |
+
"learning_rate": 0.0001773010453362737,
|
| 298 |
+
"loss": 1.3229,
|
| 299 |
+
"step": 190
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.0,
|
| 303 |
+
"eval_loss": 2.3739588260650635,
|
| 304 |
+
"eval_runtime": 0.8887,
|
| 305 |
+
"eval_samples_per_second": 6.751,
|
| 306 |
+
"eval_steps_per_second": 1.125,
|
| 307 |
+
"step": 192
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 3.046875,
|
| 311 |
+
"grad_norm": 0.13525685667991638,
|
| 312 |
+
"learning_rate": 0.00017554247804459316,
|
| 313 |
+
"loss": 1.3138,
|
| 314 |
+
"step": 195
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.125,
|
| 318 |
+
"grad_norm": 0.1221759021282196,
|
| 319 |
+
"learning_rate": 0.0001737277336810124,
|
| 320 |
+
"loss": 1.3114,
|
| 321 |
+
"step": 200
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.203125,
|
| 325 |
+
"grad_norm": 0.1508011817932129,
|
| 326 |
+
"learning_rate": 0.0001718581617779698,
|
| 327 |
+
"loss": 1.3121,
|
| 328 |
+
"step": 205
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 3.28125,
|
| 332 |
+
"grad_norm": 0.15374642610549927,
|
| 333 |
+
"learning_rate": 0.00016993515264033672,
|
| 334 |
+
"loss": 1.3061,
|
| 335 |
+
"step": 210
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 3.359375,
|
| 339 |
+
"grad_norm": 0.13054175674915314,
|
| 340 |
+
"learning_rate": 0.00016796013631151897,
|
| 341 |
+
"loss": 1.3131,
|
| 342 |
+
"step": 215
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 3.4375,
|
| 346 |
+
"grad_norm": 0.13617338240146637,
|
| 347 |
+
"learning_rate": 0.00016593458151000688,
|
| 348 |
+
"loss": 1.3055,
|
| 349 |
+
"step": 220
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 3.515625,
|
| 353 |
+
"grad_norm": 0.1574951857328415,
|
| 354 |
+
"learning_rate": 0.00016385999453716454,
|
| 355 |
+
"loss": 1.3081,
|
| 356 |
+
"step": 225
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 3.59375,
|
| 360 |
+
"grad_norm": 0.13046763837337494,
|
| 361 |
+
"learning_rate": 0.00016173791815707051,
|
| 362 |
+
"loss": 1.2971,
|
| 363 |
+
"step": 230
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.671875,
|
| 367 |
+
"grad_norm": 0.12328305840492249,
|
| 368 |
+
"learning_rate": 0.00015956993044924334,
|
| 369 |
+
"loss": 1.3004,
|
| 370 |
+
"step": 235
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 3.75,
|
| 374 |
+
"grad_norm": 0.1457005739212036,
|
| 375 |
+
"learning_rate": 0.0001573576436351046,
|
| 376 |
+
"loss": 1.301,
|
| 377 |
+
"step": 240
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 3.828125,
|
| 381 |
+
"grad_norm": 0.14775606989860535,
|
| 382 |
+
"learning_rate": 0.0001551027028790524,
|
| 383 |
+
"loss": 1.2981,
|
| 384 |
+
"step": 245
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 3.90625,
|
| 388 |
+
"grad_norm": 0.13573531806468964,
|
| 389 |
+
"learning_rate": 0.0001528067850650368,
|
| 390 |
+
"loss": 1.2964,
|
| 391 |
+
"step": 250
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 3.984375,
|
| 395 |
+
"grad_norm": 0.14599502086639404,
|
| 396 |
+
"learning_rate": 0.0001504715975495472,
|
| 397 |
+
"loss": 1.2996,
|
| 398 |
+
"step": 255
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.0,
|
| 402 |
+
"eval_loss": 2.3706400394439697,
|
| 403 |
+
"eval_runtime": 0.8905,
|
| 404 |
+
"eval_samples_per_second": 6.738,
|
| 405 |
+
"eval_steps_per_second": 1.123,
|
| 406 |
+
"step": 256
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.0625,
|
| 410 |
+
"grad_norm": 0.18352623283863068,
|
| 411 |
+
"learning_rate": 0.00014809887689193877,
|
| 412 |
+
"loss": 1.2791,
|
| 413 |
+
"step": 260
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.140625,
|
| 417 |
+
"grad_norm": 0.12809208035469055,
|
| 418 |
+
"learning_rate": 0.00014569038756304207,
|
| 419 |
+
"loss": 1.2838,
|
| 420 |
+
"step": 265
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.21875,
|
| 424 |
+
"grad_norm": 0.14866453409194946,
|
| 425 |
+
"learning_rate": 0.00014324792063301662,
|
| 426 |
+
"loss": 1.2855,
|
| 427 |
+
"step": 270
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 4.296875,
|
| 431 |
+
"grad_norm": 0.1483401656150818,
|
| 432 |
+
"learning_rate": 0.00014077329243942369,
|
| 433 |
+
"loss": 1.2906,
|
| 434 |
+
"step": 275
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.375,
|
| 438 |
+
"grad_norm": 0.1301499307155609,
|
| 439 |
+
"learning_rate": 0.000138268343236509,
|
| 440 |
+
"loss": 1.2832,
|
| 441 |
+
"step": 280
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.453125,
|
| 445 |
+
"grad_norm": 0.12229609489440918,
|
| 446 |
+
"learning_rate": 0.00013573493582670003,
|
| 447 |
+
"loss": 1.2855,
|
| 448 |
+
"step": 285
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 4.53125,
|
| 452 |
+
"grad_norm": 0.1406661719083786,
|
| 453 |
+
"learning_rate": 0.00013317495417533524,
|
| 454 |
+
"loss": 1.2885,
|
| 455 |
+
"step": 290
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 4.609375,
|
| 459 |
+
"grad_norm": 0.13113024830818176,
|
| 460 |
+
"learning_rate": 0.00013059030200965536,
|
| 461 |
+
"loss": 1.2899,
|
| 462 |
+
"step": 295
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 4.6875,
|
| 466 |
+
"grad_norm": 0.13431058824062347,
|
| 467 |
+
"learning_rate": 0.00012798290140309923,
|
| 468 |
+
"loss": 1.2819,
|
| 469 |
+
"step": 300
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 4.765625,
|
| 473 |
+
"grad_norm": 0.14978773891925812,
|
| 474 |
+
"learning_rate": 0.00012535469134595595,
|
| 475 |
+
"loss": 1.2852,
|
| 476 |
+
"step": 305
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 4.84375,
|
| 480 |
+
"grad_norm": 0.12423586845397949,
|
| 481 |
+
"learning_rate": 0.00012270762630343734,
|
| 482 |
+
"loss": 1.2802,
|
| 483 |
+
"step": 310
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 4.921875,
|
| 487 |
+
"grad_norm": 0.14971400797367096,
|
| 488 |
+
"learning_rate": 0.00012004367476224206,
|
| 489 |
+
"loss": 1.2829,
|
| 490 |
+
"step": 315
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 5.0,
|
| 494 |
+
"grad_norm": 0.14705318212509155,
|
| 495 |
+
"learning_rate": 0.00011736481776669306,
|
| 496 |
+
"loss": 1.2878,
|
| 497 |
+
"step": 320
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 5.0,
|
| 501 |
+
"eval_loss": 2.3755218982696533,
|
| 502 |
+
"eval_runtime": 0.8842,
|
| 503 |
+
"eval_samples_per_second": 6.786,
|
| 504 |
+
"eval_steps_per_second": 1.131,
|
| 505 |
+
"step": 320
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.078125,
|
| 509 |
+
"grad_norm": 0.17323575913906097,
|
| 510 |
+
"learning_rate": 0.00011467304744553618,
|
| 511 |
+
"loss": 1.2791,
|
| 512 |
+
"step": 325
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.15625,
|
| 516 |
+
"grad_norm": 0.15446850657463074,
|
| 517 |
+
"learning_rate": 0.00011197036553049625,
|
| 518 |
+
"loss": 1.2663,
|
| 519 |
+
"step": 330
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.234375,
|
| 523 |
+
"grad_norm": 0.1357167363166809,
|
| 524 |
+
"learning_rate": 0.00010925878186769158,
|
| 525 |
+
"loss": 1.2707,
|
| 526 |
+
"step": 335
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.3125,
|
| 530 |
+
"grad_norm": 0.13887612521648407,
|
| 531 |
+
"learning_rate": 0.00010654031292301432,
|
| 532 |
+
"loss": 1.2713,
|
| 533 |
+
"step": 340
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 5.390625,
|
| 537 |
+
"grad_norm": 0.18401511013507843,
|
| 538 |
+
"learning_rate": 0.00010381698028258817,
|
| 539 |
+
"loss": 1.2708,
|
| 540 |
+
"step": 345
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 5.46875,
|
| 544 |
+
"grad_norm": 0.14264121651649475,
|
| 545 |
+
"learning_rate": 0.00010109080914941824,
|
| 546 |
+
"loss": 1.2716,
|
| 547 |
+
"step": 350
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 5.546875,
|
| 551 |
+
"grad_norm": 0.1242898479104042,
|
| 552 |
+
"learning_rate": 9.836382683735132e-05,
|
| 553 |
+
"loss": 1.2778,
|
| 554 |
+
"step": 355
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 5.625,
|
| 558 |
+
"grad_norm": 0.11967791616916656,
|
| 559 |
+
"learning_rate": 9.563806126346642e-05,
|
| 560 |
+
"loss": 1.2761,
|
| 561 |
+
"step": 360
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 5.703125,
|
| 565 |
+
"grad_norm": 0.1407729834318161,
|
| 566 |
+
"learning_rate": 9.29155394400166e-05,
|
| 567 |
+
"loss": 1.271,
|
| 568 |
+
"step": 365
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 5.78125,
|
| 572 |
+
"grad_norm": 0.127610981464386,
|
| 573 |
+
"learning_rate": 9.019828596704394e-05,
|
| 574 |
+
"loss": 1.2707,
|
| 575 |
+
"step": 370
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 5.859375,
|
| 579 |
+
"grad_norm": 0.131484255194664,
|
| 580 |
+
"learning_rate": 8.74883215267881e-05,
|
| 581 |
+
"loss": 1.2703,
|
| 582 |
+
"step": 375
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 5.9375,
|
| 586 |
+
"grad_norm": 0.14869672060012817,
|
| 587 |
+
"learning_rate": 8.478766138100834e-05,
|
| 588 |
+
"loss": 1.2746,
|
| 589 |
+
"step": 380
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 6.0,
|
| 593 |
+
"eval_loss": 2.3722288608551025,
|
| 594 |
+
"eval_runtime": 0.8691,
|
| 595 |
+
"eval_samples_per_second": 6.904,
|
| 596 |
+
"eval_steps_per_second": 1.151,
|
| 597 |
+
"step": 384
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 6.015625,
|
| 601 |
+
"grad_norm": 0.12939053773880005,
|
| 602 |
+
"learning_rate": 8.209831387233676e-05,
|
| 603 |
+
"loss": 1.267,
|
| 604 |
+
"step": 385
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 6.09375,
|
| 608 |
+
"grad_norm": 0.1274917721748352,
|
| 609 |
+
"learning_rate": 7.942227893077652e-05,
|
| 610 |
+
"loss": 1.2609,
|
| 611 |
+
"step": 390
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 6.171875,
|
| 615 |
+
"grad_norm": 0.128373920917511,
|
| 616 |
+
"learning_rate": 7.676154658645656e-05,
|
| 617 |
+
"loss": 1.2613,
|
| 618 |
+
"step": 395
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 6.25,
|
| 622 |
+
"grad_norm": 0.12212257087230682,
|
| 623 |
+
"learning_rate": 7.411809548974792e-05,
|
| 624 |
+
"loss": 1.2608,
|
| 625 |
+
"step": 400
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 6.328125,
|
| 629 |
+
"grad_norm": 0.12020603567361832,
|
| 630 |
+
"learning_rate": 7.149389143984295e-05,
|
| 631 |
+
"loss": 1.2648,
|
| 632 |
+
"step": 405
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 6.40625,
|
| 636 |
+
"grad_norm": 0.13269896805286407,
|
| 637 |
+
"learning_rate": 6.889088592289093e-05,
|
| 638 |
+
"loss": 1.263,
|
| 639 |
+
"step": 410
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 6.484375,
|
| 643 |
+
"grad_norm": 0.12962587177753448,
|
| 644 |
+
"learning_rate": 6.6311014660778e-05,
|
| 645 |
+
"loss": 1.2688,
|
| 646 |
+
"step": 415
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 6.5625,
|
| 650 |
+
"grad_norm": 0.13470596075057983,
|
| 651 |
+
"learning_rate": 6.375619617162985e-05,
|
| 652 |
+
"loss": 1.2606,
|
| 653 |
+
"step": 420
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 6.640625,
|
| 657 |
+
"grad_norm": 0.1306695193052292,
|
| 658 |
+
"learning_rate": 6.122833034310793e-05,
|
| 659 |
+
"loss": 1.264,
|
| 660 |
+
"step": 425
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 6.71875,
|
| 664 |
+
"grad_norm": 0.12205986678600311,
|
| 665 |
+
"learning_rate": 5.872929701956054e-05,
|
| 666 |
+
"loss": 1.2624,
|
| 667 |
+
"step": 430
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 6.796875,
|
| 671 |
+
"grad_norm": 0.12223726511001587,
|
| 672 |
+
"learning_rate": 5.6260954604078585e-05,
|
| 673 |
+
"loss": 1.2624,
|
| 674 |
+
"step": 435
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 6.875,
|
| 678 |
+
"grad_norm": 0.11978649348020554,
|
| 679 |
+
"learning_rate": 5.382513867649663e-05,
|
| 680 |
+
"loss": 1.2684,
|
| 681 |
+
"step": 440
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 6.953125,
|
| 685 |
+
"grad_norm": 0.13118350505828857,
|
| 686 |
+
"learning_rate": 5.142366062836599e-05,
|
| 687 |
+
"loss": 1.2617,
|
| 688 |
+
"step": 445
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 7.0,
|
| 692 |
+
"eval_loss": 2.3756062984466553,
|
| 693 |
+
"eval_runtime": 0.8698,
|
| 694 |
+
"eval_samples_per_second": 6.898,
|
| 695 |
+
"eval_steps_per_second": 1.15,
|
| 696 |
+
"step": 448
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 7.03125,
|
| 700 |
+
"grad_norm": 0.12501190602779388,
|
| 701 |
+
"learning_rate": 4.9058306315915826e-05,
|
| 702 |
+
"loss": 1.2686,
|
| 703 |
+
"step": 450
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 7.109375,
|
| 707 |
+
"grad_norm": 0.11916686594486237,
|
| 708 |
+
"learning_rate": 4.6730834732003104e-05,
|
| 709 |
+
"loss": 1.2589,
|
| 710 |
+
"step": 455
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 7.1875,
|
| 714 |
+
"grad_norm": 0.1307828575372696,
|
| 715 |
+
"learning_rate": 4.444297669803981e-05,
|
| 716 |
+
"loss": 1.2641,
|
| 717 |
+
"step": 460
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 7.265625,
|
| 721 |
+
"grad_norm": 0.1322871744632721,
|
| 722 |
+
"learning_rate": 4.219643357686967e-05,
|
| 723 |
+
"loss": 1.2637,
|
| 724 |
+
"step": 465
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 7.34375,
|
| 728 |
+
"grad_norm": 0.12456091493368149,
|
| 729 |
+
"learning_rate": 3.999287600755192e-05,
|
| 730 |
+
"loss": 1.2596,
|
| 731 |
+
"step": 470
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 7.421875,
|
| 735 |
+
"grad_norm": 0.12441947311162949,
|
| 736 |
+
"learning_rate": 3.783394266299228e-05,
|
| 737 |
+
"loss": 1.2609,
|
| 738 |
+
"step": 475
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 7.5,
|
| 742 |
+
"grad_norm": 0.1233881339430809,
|
| 743 |
+
"learning_rate": 3.5721239031346066e-05,
|
| 744 |
+
"loss": 1.258,
|
| 745 |
+
"step": 480
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 7.578125,
|
| 749 |
+
"grad_norm": 0.11959797888994217,
|
| 750 |
+
"learning_rate": 3.365633622209891e-05,
|
| 751 |
+
"loss": 1.2535,
|
| 752 |
+
"step": 485
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 7.65625,
|
| 756 |
+
"grad_norm": 0.13552451133728027,
|
| 757 |
+
"learning_rate": 3.164076979771287e-05,
|
| 758 |
+
"loss": 1.2567,
|
| 759 |
+
"step": 490
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 7.734375,
|
| 763 |
+
"grad_norm": 0.1312321126461029,
|
| 764 |
+
"learning_rate": 2.9676038631707593e-05,
|
| 765 |
+
"loss": 1.2546,
|
| 766 |
+
"step": 495
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 7.8125,
|
| 770 |
+
"grad_norm": 0.12341451644897461,
|
| 771 |
+
"learning_rate": 2.776360379402445e-05,
|
| 772 |
+
"loss": 1.2589,
|
| 773 |
+
"step": 500
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 7.890625,
|
| 777 |
+
"grad_norm": 0.11580604314804077,
|
| 778 |
+
"learning_rate": 2.5904887464504114e-05,
|
| 779 |
+
"loss": 1.2514,
|
| 780 |
+
"step": 505
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 7.96875,
|
| 784 |
+
"grad_norm": 0.1164306178689003,
|
| 785 |
+
"learning_rate": 2.4101271875283817e-05,
|
| 786 |
+
"loss": 1.2497,
|
| 787 |
+
"step": 510
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 8.0,
|
| 791 |
+
"eval_loss": 2.3754327297210693,
|
| 792 |
+
"eval_runtime": 0.8715,
|
| 793 |
+
"eval_samples_per_second": 6.885,
|
| 794 |
+
"eval_steps_per_second": 1.147,
|
| 795 |
+
"step": 512
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 8.046875,
|
| 799 |
+
"grad_norm": 0.11353053152561188,
|
| 800 |
+
"learning_rate": 2.2354098282902446e-05,
|
| 801 |
+
"loss": 1.2515,
|
| 802 |
+
"step": 515
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 8.125,
|
| 806 |
+
"grad_norm": 0.11808107793331146,
|
| 807 |
+
"learning_rate": 2.0664665970876496e-05,
|
| 808 |
+
"loss": 1.2568,
|
| 809 |
+
"step": 520
|
| 810 |
+
},
|
| 811 |
+
{
|
| 812 |
+
"epoch": 8.203125,
|
| 813 |
+
"grad_norm": 0.12227106839418411,
|
| 814 |
+
"learning_rate": 1.903423128348959e-05,
|
| 815 |
+
"loss": 1.2526,
|
| 816 |
+
"step": 525
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 8.28125,
|
| 820 |
+
"grad_norm": 0.11858697980642319,
|
| 821 |
+
"learning_rate": 1.7464006691513623e-05,
|
| 822 |
+
"loss": 1.2546,
|
| 823 |
+
"step": 530
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 8.359375,
|
| 827 |
+
"grad_norm": 0.11320216953754425,
|
| 828 |
+
"learning_rate": 1.595515989055618e-05,
|
| 829 |
+
"loss": 1.2513,
|
| 830 |
+
"step": 535
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 8.4375,
|
| 834 |
+
"grad_norm": 0.11540042608976364,
|
| 835 |
+
"learning_rate": 1.4508812932705363e-05,
|
| 836 |
+
"loss": 1.2539,
|
| 837 |
+
"step": 540
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 8.515625,
|
| 841 |
+
"grad_norm": 0.11582642793655396,
|
| 842 |
+
"learning_rate": 1.3126041392116772e-05,
|
| 843 |
+
"loss": 1.2552,
|
| 844 |
+
"step": 545
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 8.59375,
|
| 848 |
+
"grad_norm": 0.11406353861093521,
|
| 849 |
+
"learning_rate": 1.1807873565164506e-05,
|
| 850 |
+
"loss": 1.2451,
|
| 851 |
+
"step": 550
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 8.671875,
|
| 855 |
+
"grad_norm": 0.11722878366708755,
|
| 856 |
+
"learning_rate": 1.0555289705749483e-05,
|
| 857 |
+
"loss": 1.2555,
|
| 858 |
+
"step": 555
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 8.75,
|
| 862 |
+
"grad_norm": 0.11788377165794373,
|
| 863 |
+
"learning_rate": 9.369221296335006e-06,
|
| 864 |
+
"loss": 1.2644,
|
| 865 |
+
"step": 560
|
| 866 |
+
},
|
| 867 |
+
{
|
| 868 |
+
"epoch": 8.828125,
|
| 869 |
+
"grad_norm": 0.12228409945964813,
|
| 870 |
+
"learning_rate": 8.250550355250875e-06,
|
| 871 |
+
"loss": 1.2554,
|
| 872 |
+
"step": 565
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 8.90625,
|
| 876 |
+
"grad_norm": 0.11590871214866638,
|
| 877 |
+
"learning_rate": 7.200108780781556e-06,
|
| 878 |
+
"loss": 1.2581,
|
| 879 |
+
"step": 570
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 8.984375,
|
| 883 |
+
"grad_norm": 0.11986227333545685,
|
| 884 |
+
"learning_rate": 6.218677732526035e-06,
|
| 885 |
+
"loss": 1.2549,
|
| 886 |
+
"step": 575
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 9.0,
|
| 890 |
+
"eval_loss": 2.3762052059173584,
|
| 891 |
+
"eval_runtime": 0.8701,
|
| 892 |
+
"eval_samples_per_second": 6.896,
|
| 893 |
+
"eval_steps_per_second": 1.149,
|
| 894 |
+
"step": 576
|
| 895 |
+
},
|
| 896 |
+
{
|
| 897 |
+
"epoch": 9.0625,
|
| 898 |
+
"grad_norm": 0.11538632959127426,
|
| 899 |
+
"learning_rate": 5.306987050489442e-06,
|
| 900 |
+
"loss": 1.2553,
|
| 901 |
+
"step": 580
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 9.140625,
|
| 905 |
+
"grad_norm": 0.11359097808599472,
|
| 906 |
+
"learning_rate": 4.465714712338398e-06,
|
| 907 |
+
"loss": 1.2438,
|
| 908 |
+
"step": 585
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 9.21875,
|
| 912 |
+
"grad_norm": 0.1167021319270134,
|
| 913 |
+
"learning_rate": 3.6954863292237297e-06,
|
| 914 |
+
"loss": 1.2508,
|
| 915 |
+
"step": 590
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 9.296875,
|
| 919 |
+
"grad_norm": 0.11401596665382385,
|
| 920 |
+
"learning_rate": 2.996874680545603e-06,
|
| 921 |
+
"loss": 1.2555,
|
| 922 |
+
"step": 595
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 9.375,
|
| 926 |
+
"grad_norm": 0.11593978106975555,
|
| 927 |
+
"learning_rate": 2.3703992880066638e-06,
|
| 928 |
+
"loss": 1.2491,
|
| 929 |
+
"step": 600
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 9.453125,
|
| 933 |
+
"grad_norm": 0.11369957774877548,
|
| 934 |
+
"learning_rate": 1.8165260292704711e-06,
|
| 935 |
+
"loss": 1.2556,
|
| 936 |
+
"step": 605
|
| 937 |
+
},
|
| 938 |
+
{
|
| 939 |
+
"epoch": 9.53125,
|
| 940 |
+
"grad_norm": 0.11447325348854065,
|
| 941 |
+
"learning_rate": 1.3356667915121025e-06,
|
| 942 |
+
"loss": 1.2559,
|
| 943 |
+
"step": 610
|
| 944 |
+
},
|
| 945 |
+
{
|
| 946 |
+
"epoch": 9.609375,
|
| 947 |
+
"grad_norm": 0.11217474192380905,
|
| 948 |
+
"learning_rate": 9.281791651187366e-07,
|
| 949 |
+
"loss": 1.2635,
|
| 950 |
+
"step": 615
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 9.6875,
|
| 954 |
+
"grad_norm": 0.11821179836988449,
|
| 955 |
+
"learning_rate": 5.943661777680354e-07,
|
| 956 |
+
"loss": 1.2521,
|
| 957 |
+
"step": 620
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 9.765625,
|
| 961 |
+
"grad_norm": 0.113133005797863,
|
| 962 |
+
"learning_rate": 3.3447606908196817e-07,
|
| 963 |
+
"loss": 1.2539,
|
| 964 |
+
"step": 625
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 9.84375,
|
| 968 |
+
"grad_norm": 0.11367449164390564,
|
| 969 |
+
"learning_rate": 1.487021060236904e-07,
|
| 970 |
+
"loss": 1.2528,
|
| 971 |
+
"step": 630
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 9.921875,
|
| 975 |
+
"grad_norm": 0.11390256881713867,
|
| 976 |
+
"learning_rate": 3.7182439174832106e-08,
|
| 977 |
+
"loss": 1.2581,
|
| 978 |
+
"step": 635
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 10.0,
|
| 982 |
+
"grad_norm": 0.11366365104913712,
|
| 983 |
+
"learning_rate": 0.0,
|
| 984 |
+
"loss": 1.2494,
|
| 985 |
+
"step": 640
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 10.0,
|
| 989 |
+
"eval_loss": 2.376882791519165,
|
| 990 |
+
"eval_runtime": 0.8974,
|
| 991 |
+
"eval_samples_per_second": 6.686,
|
| 992 |
+
"eval_steps_per_second": 1.114,
|
| 993 |
+
"step": 640
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 10.0,
|
| 997 |
+
"step": 640,
|
| 998 |
+
"total_flos": 2.846679860191953e+18,
|
| 999 |
+
"train_loss": 1.3252380434423685,
|
| 1000 |
+
"train_runtime": 3360.483,
|
| 1001 |
+
"train_samples_per_second": 48.669,
|
| 1002 |
+
"train_steps_per_second": 0.19
|
| 1003 |
+
}
|
| 1004 |
+
],
|
| 1005 |
+
"logging_steps": 5,
|
| 1006 |
+
"max_steps": 640,
|
| 1007 |
+
"num_input_tokens_seen": 0,
|
| 1008 |
+
"num_train_epochs": 10,
|
| 1009 |
+
"save_steps": 100,
|
| 1010 |
+
"stateful_callbacks": {
|
| 1011 |
+
"TrainerControl": {
|
| 1012 |
+
"args": {
|
| 1013 |
+
"should_epoch_stop": false,
|
| 1014 |
+
"should_evaluate": false,
|
| 1015 |
+
"should_log": false,
|
| 1016 |
+
"should_save": true,
|
| 1017 |
+
"should_training_stop": true
|
| 1018 |
+
},
|
| 1019 |
+
"attributes": {}
|
| 1020 |
+
}
|
| 1021 |
+
},
|
| 1022 |
+
"total_flos": 2.846679860191953e+18,
|
| 1023 |
+
"train_batch_size": 16,
|
| 1024 |
+
"trial_name": null,
|
| 1025 |
+
"trial_params": null
|
| 1026 |
+
}
|