Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,66 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: llama3.1
|
| 4 |
+
base_model: meta-llama/Meta-Llama-3.1-8B-Instruct
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: barc-llama3.1-8b-instruct-fft-transduction-gpt4omini10k_lr1e-5_epoch1_
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# barc-llama3.1-8b-instruct-fft-transduction-gpt4omini10k_lr1e-5_epoch1_
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B-Instruct) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.7619
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 1e-05
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 8
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 128
|
| 48 |
+
- total_eval_batch_size: 32
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: cosine
|
| 51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
+
- num_epochs: 1
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 58 |
+
| 0.8604 | 1.0 | 76 | 0.7619 |
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
### Framework versions
|
| 62 |
+
|
| 63 |
+
- Transformers 4.45.0.dev0
|
| 64 |
+
- Pytorch 2.4.1+cu121
|
| 65 |
+
- Datasets 3.0.0
|
| 66 |
+
- Tokenizers 0.20.0
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 2.244031216091136e+16,
|
| 4 |
+
"train_loss": 2.2036053005017733,
|
| 5 |
+
"train_runtime": 949.5806,
|
| 6 |
+
"train_samples": 9698,
|
| 7 |
+
"train_samples_per_second": 10.213,
|
| 8 |
+
"train_steps_per_second": 0.08
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 128000,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
128001,
|
| 6 |
+
128008,
|
| 7 |
+
128009
|
| 8 |
+
],
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "4.45.0.dev0"
|
| 12 |
+
}
|
runs/Sep25_15-15-42_instance-20240925-140811/events.out.tfevents.1727277425.instance-20240925-140811.23739.0
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:d24bda1d972ce84a776f1a2947ed074df42184dbb6902dca73910065d4965c06
|
| 3 |
+
size 22414
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 2.244031216091136e+16,
|
| 4 |
+
"train_loss": 2.2036053005017733,
|
| 5 |
+
"train_runtime": 949.5806,
|
| 6 |
+
"train_samples": 9698,
|
| 7 |
+
"train_samples_per_second": 10.213,
|
| 8 |
+
"train_steps_per_second": 0.08
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,582 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 76,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.013157894736842105,
|
| 13 |
+
"grad_norm": 1699.4203999374324,
|
| 14 |
+
"learning_rate": 1.25e-06,
|
| 15 |
+
"loss": 11.6905,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02631578947368421,
|
| 20 |
+
"grad_norm": 1649.6649286698323,
|
| 21 |
+
"learning_rate": 2.5e-06,
|
| 22 |
+
"loss": 11.7067,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.039473684210526314,
|
| 27 |
+
"grad_norm": 1109.6668329617996,
|
| 28 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 29 |
+
"loss": 10.8238,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.05263157894736842,
|
| 34 |
+
"grad_norm": 1106.178486585862,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 9.0694,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.06578947368421052,
|
| 41 |
+
"grad_norm": 809.3294910843833,
|
| 42 |
+
"learning_rate": 6.25e-06,
|
| 43 |
+
"loss": 5.9203,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.07894736842105263,
|
| 48 |
+
"grad_norm": 257.0422697393994,
|
| 49 |
+
"learning_rate": 7.500000000000001e-06,
|
| 50 |
+
"loss": 4.7981,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.09210526315789473,
|
| 55 |
+
"grad_norm": 394.7803420902946,
|
| 56 |
+
"learning_rate": 8.750000000000001e-06,
|
| 57 |
+
"loss": 5.5132,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.10526315789473684,
|
| 62 |
+
"grad_norm": 126.60762803235828,
|
| 63 |
+
"learning_rate": 1e-05,
|
| 64 |
+
"loss": 3.2079,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.11842105263157894,
|
| 69 |
+
"grad_norm": 125.30453375076969,
|
| 70 |
+
"learning_rate": 9.994664874011864e-06,
|
| 71 |
+
"loss": 5.0667,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.13157894736842105,
|
| 76 |
+
"grad_norm": 158.38384375765978,
|
| 77 |
+
"learning_rate": 9.978670881475173e-06,
|
| 78 |
+
"loss": 4.1611,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.14473684210526316,
|
| 83 |
+
"grad_norm": 55.659050974987274,
|
| 84 |
+
"learning_rate": 9.952052154376027e-06,
|
| 85 |
+
"loss": 3.4048,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.15789473684210525,
|
| 90 |
+
"grad_norm": 42.873598431489626,
|
| 91 |
+
"learning_rate": 9.91486549841951e-06,
|
| 92 |
+
"loss": 4.0195,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.17105263157894737,
|
| 97 |
+
"grad_norm": 72.14863137487674,
|
| 98 |
+
"learning_rate": 9.867190271803466e-06,
|
| 99 |
+
"loss": 4.0778,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.18421052631578946,
|
| 104 |
+
"grad_norm": 39.55596624092672,
|
| 105 |
+
"learning_rate": 9.809128215864096e-06,
|
| 106 |
+
"loss": 3.8024,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.19736842105263158,
|
| 111 |
+
"grad_norm": 18.04964705613809,
|
| 112 |
+
"learning_rate": 9.74080323795483e-06,
|
| 113 |
+
"loss": 3.3671,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.21052631578947367,
|
| 118 |
+
"grad_norm": 14.28948406963546,
|
| 119 |
+
"learning_rate": 9.66236114702178e-06,
|
| 120 |
+
"loss": 2.7161,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2236842105263158,
|
| 125 |
+
"grad_norm": 14.98902092554103,
|
| 126 |
+
"learning_rate": 9.573969342440107e-06,
|
| 127 |
+
"loss": 2.6427,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.23684210526315788,
|
| 132 |
+
"grad_norm": 17.857073526805326,
|
| 133 |
+
"learning_rate": 9.475816456775313e-06,
|
| 134 |
+
"loss": 2.5664,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.25,
|
| 139 |
+
"grad_norm": 17.982906576778536,
|
| 140 |
+
"learning_rate": 9.368111953231849e-06,
|
| 141 |
+
"loss": 2.3566,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.2631578947368421,
|
| 146 |
+
"grad_norm": 14.066497831682323,
|
| 147 |
+
"learning_rate": 9.251085678648072e-06,
|
| 148 |
+
"loss": 2.3749,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.27631578947368424,
|
| 153 |
+
"grad_norm": 16.568001275060062,
|
| 154 |
+
"learning_rate": 9.124987372991512e-06,
|
| 155 |
+
"loss": 2.3824,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2894736842105263,
|
| 160 |
+
"grad_norm": 10.225207891248253,
|
| 161 |
+
"learning_rate": 8.990086136401199e-06,
|
| 162 |
+
"loss": 2.2876,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3026315789473684,
|
| 167 |
+
"grad_norm": 15.52475147772328,
|
| 168 |
+
"learning_rate": 8.846669854914395e-06,
|
| 169 |
+
"loss": 2.0579,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3157894736842105,
|
| 174 |
+
"grad_norm": 10.735876859280078,
|
| 175 |
+
"learning_rate": 8.695044586103297e-06,
|
| 176 |
+
"loss": 2.1149,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.32894736842105265,
|
| 181 |
+
"grad_norm": 12.03589457222836,
|
| 182 |
+
"learning_rate": 8.535533905932739e-06,
|
| 183 |
+
"loss": 2.0015,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.34210526315789475,
|
| 188 |
+
"grad_norm": 10.668355704543856,
|
| 189 |
+
"learning_rate": 8.368478218232787e-06,
|
| 190 |
+
"loss": 2.0109,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.35526315789473684,
|
| 195 |
+
"grad_norm": 11.247848100294151,
|
| 196 |
+
"learning_rate": 8.194234028259806e-06,
|
| 197 |
+
"loss": 2.0085,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.3684210526315789,
|
| 202 |
+
"grad_norm": 10.041875020749039,
|
| 203 |
+
"learning_rate": 8.013173181896283e-06,
|
| 204 |
+
"loss": 1.707,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3815789473684211,
|
| 209 |
+
"grad_norm": 16.66505157922524,
|
| 210 |
+
"learning_rate": 7.82568207211296e-06,
|
| 211 |
+
"loss": 1.9144,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.39473684210526316,
|
| 216 |
+
"grad_norm": 14.129248059213934,
|
| 217 |
+
"learning_rate": 7.63216081438678e-06,
|
| 218 |
+
"loss": 1.7373,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.40789473684210525,
|
| 223 |
+
"grad_norm": 8.400365715854278,
|
| 224 |
+
"learning_rate": 7.4330223928342814e-06,
|
| 225 |
+
"loss": 1.606,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.42105263157894735,
|
| 230 |
+
"grad_norm": 9.454647164990588,
|
| 231 |
+
"learning_rate": 7.2286917788826926e-06,
|
| 232 |
+
"loss": 1.5251,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4342105263157895,
|
| 237 |
+
"grad_norm": 11.708573459019293,
|
| 238 |
+
"learning_rate": 7.019605024359475e-06,
|
| 239 |
+
"loss": 1.4462,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.4473684210526316,
|
| 244 |
+
"grad_norm": 11.288211975525503,
|
| 245 |
+
"learning_rate": 6.806208330935766e-06,
|
| 246 |
+
"loss": 1.3242,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.4605263157894737,
|
| 251 |
+
"grad_norm": 12.84833845270528,
|
| 252 |
+
"learning_rate": 6.588957097909509e-06,
|
| 253 |
+
"loss": 1.2777,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.47368421052631576,
|
| 258 |
+
"grad_norm": 12.805717587983919,
|
| 259 |
+
"learning_rate": 6.368314950360416e-06,
|
| 260 |
+
"loss": 1.2439,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4868421052631579,
|
| 265 |
+
"grad_norm": 16.79814160674569,
|
| 266 |
+
"learning_rate": 6.144752749750671e-06,
|
| 267 |
+
"loss": 1.2062,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.5,
|
| 272 |
+
"grad_norm": 11.704864343239466,
|
| 273 |
+
"learning_rate": 5.918747589082853e-06,
|
| 274 |
+
"loss": 1.2953,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5131578947368421,
|
| 279 |
+
"grad_norm": 9.088068501685061,
|
| 280 |
+
"learning_rate": 5.690781774759412e-06,
|
| 281 |
+
"loss": 1.1413,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5263157894736842,
|
| 286 |
+
"grad_norm": 9.452972340503731,
|
| 287 |
+
"learning_rate": 5.46134179731651e-06,
|
| 288 |
+
"loss": 1.1089,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5394736842105263,
|
| 293 |
+
"grad_norm": 13.866484586191572,
|
| 294 |
+
"learning_rate": 5.230917293228699e-06,
|
| 295 |
+
"loss": 1.0386,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5526315789473685,
|
| 300 |
+
"grad_norm": 9.961380544449332,
|
| 301 |
+
"learning_rate": 5e-06,
|
| 302 |
+
"loss": 1.0679,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5657894736842105,
|
| 307 |
+
"grad_norm": 8.460862397122703,
|
| 308 |
+
"learning_rate": 4.7690827067713035e-06,
|
| 309 |
+
"loss": 1.1567,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5789473684210527,
|
| 314 |
+
"grad_norm": 8.823619822316715,
|
| 315 |
+
"learning_rate": 4.53865820268349e-06,
|
| 316 |
+
"loss": 1.049,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.5921052631578947,
|
| 321 |
+
"grad_norm": 7.271462434248026,
|
| 322 |
+
"learning_rate": 4.309218225240591e-06,
|
| 323 |
+
"loss": 0.9448,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6052631578947368,
|
| 328 |
+
"grad_norm": 15.732461239193407,
|
| 329 |
+
"learning_rate": 4.081252410917148e-06,
|
| 330 |
+
"loss": 0.9914,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.618421052631579,
|
| 335 |
+
"grad_norm": 11.654049140402229,
|
| 336 |
+
"learning_rate": 3.855247250249331e-06,
|
| 337 |
+
"loss": 0.8027,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.631578947368421,
|
| 342 |
+
"grad_norm": 17.59370496573567,
|
| 343 |
+
"learning_rate": 3.6316850496395863e-06,
|
| 344 |
+
"loss": 1.0106,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6447368421052632,
|
| 349 |
+
"grad_norm": 10.035162483746694,
|
| 350 |
+
"learning_rate": 3.4110429020904924e-06,
|
| 351 |
+
"loss": 0.8962,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.6578947368421053,
|
| 356 |
+
"grad_norm": 7.517763142577863,
|
| 357 |
+
"learning_rate": 3.1937916690642356e-06,
|
| 358 |
+
"loss": 0.9492,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6710526315789473,
|
| 363 |
+
"grad_norm": 15.969988377607551,
|
| 364 |
+
"learning_rate": 2.980394975640526e-06,
|
| 365 |
+
"loss": 0.8697,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6842105263157895,
|
| 370 |
+
"grad_norm": 7.9971953326811125,
|
| 371 |
+
"learning_rate": 2.771308221117309e-06,
|
| 372 |
+
"loss": 0.9477,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6973684210526315,
|
| 377 |
+
"grad_norm": 9.166442366719266,
|
| 378 |
+
"learning_rate": 2.5669776071657194e-06,
|
| 379 |
+
"loss": 0.9585,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.7105263157894737,
|
| 384 |
+
"grad_norm": 8.283342586733458,
|
| 385 |
+
"learning_rate": 2.3678391856132203e-06,
|
| 386 |
+
"loss": 0.948,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.7236842105263158,
|
| 391 |
+
"grad_norm": 9.135125773249728,
|
| 392 |
+
"learning_rate": 2.174317927887041e-06,
|
| 393 |
+
"loss": 0.8435,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7368421052631579,
|
| 398 |
+
"grad_norm": 7.738718289232403,
|
| 399 |
+
"learning_rate": 1.9868268181037186e-06,
|
| 400 |
+
"loss": 0.9933,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.75,
|
| 405 |
+
"grad_norm": 5.632776555168388,
|
| 406 |
+
"learning_rate": 1.8057659717401948e-06,
|
| 407 |
+
"loss": 0.8011,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7631578947368421,
|
| 412 |
+
"grad_norm": 6.642349176520887,
|
| 413 |
+
"learning_rate": 1.6315217817672142e-06,
|
| 414 |
+
"loss": 0.7914,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7763157894736842,
|
| 419 |
+
"grad_norm": 6.204837918911468,
|
| 420 |
+
"learning_rate": 1.4644660940672628e-06,
|
| 421 |
+
"loss": 0.7612,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.7894736842105263,
|
| 426 |
+
"grad_norm": 6.881328834033968,
|
| 427 |
+
"learning_rate": 1.3049554138967052e-06,
|
| 428 |
+
"loss": 0.7503,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8026315789473685,
|
| 433 |
+
"grad_norm": 6.131564640099481,
|
| 434 |
+
"learning_rate": 1.1533301450856054e-06,
|
| 435 |
+
"loss": 0.817,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.8157894736842105,
|
| 440 |
+
"grad_norm": 5.8740721419365585,
|
| 441 |
+
"learning_rate": 1.0099138635988026e-06,
|
| 442 |
+
"loss": 0.7754,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8289473684210527,
|
| 447 |
+
"grad_norm": 5.434760658098561,
|
| 448 |
+
"learning_rate": 8.750126270084891e-07,
|
| 449 |
+
"loss": 0.8402,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8421052631578947,
|
| 454 |
+
"grad_norm": 5.408440691228196,
|
| 455 |
+
"learning_rate": 7.489143213519301e-07,
|
| 456 |
+
"loss": 0.7445,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.8552631578947368,
|
| 461 |
+
"grad_norm": 4.506252846496264,
|
| 462 |
+
"learning_rate": 6.318880467681527e-07,
|
| 463 |
+
"loss": 0.6984,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.868421052631579,
|
| 468 |
+
"grad_norm": 7.264506467869279,
|
| 469 |
+
"learning_rate": 5.241835432246888e-07,
|
| 470 |
+
"loss": 0.7549,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.881578947368421,
|
| 475 |
+
"grad_norm": 5.671765225216593,
|
| 476 |
+
"learning_rate": 4.2603065755989493e-07,
|
| 477 |
+
"loss": 0.8249,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8947368421052632,
|
| 482 |
+
"grad_norm": 4.946461980264793,
|
| 483 |
+
"learning_rate": 3.3763885297822153e-07,
|
| 484 |
+
"loss": 0.7372,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.9078947368421053,
|
| 489 |
+
"grad_norm": 7.216022490434719,
|
| 490 |
+
"learning_rate": 2.5919676204517073e-07,
|
| 491 |
+
"loss": 0.7816,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.9210526315789473,
|
| 496 |
+
"grad_norm": 4.627827587948695,
|
| 497 |
+
"learning_rate": 1.908717841359048e-07,
|
| 498 |
+
"loss": 0.786,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.9342105263157895,
|
| 503 |
+
"grad_norm": 3.6873839890732363,
|
| 504 |
+
"learning_rate": 1.328097281965357e-07,
|
| 505 |
+
"loss": 0.7109,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9473684210526315,
|
| 510 |
+
"grad_norm": 3.9116965099593646,
|
| 511 |
+
"learning_rate": 8.513450158049109e-08,
|
| 512 |
+
"loss": 0.7468,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9605263157894737,
|
| 517 |
+
"grad_norm": 4.8726672621846365,
|
| 518 |
+
"learning_rate": 4.794784562397459e-08,
|
| 519 |
+
"loss": 0.6888,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9736842105263158,
|
| 524 |
+
"grad_norm": 4.616871357784515,
|
| 525 |
+
"learning_rate": 2.1329118524827662e-08,
|
| 526 |
+
"loss": 0.7657,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.9868421052631579,
|
| 531 |
+
"grad_norm": 4.101276371849697,
|
| 532 |
+
"learning_rate": 5.3351259881379016e-09,
|
| 533 |
+
"loss": 0.686,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.0,
|
| 538 |
+
"grad_norm": 4.36368396992744,
|
| 539 |
+
"learning_rate": 0.0,
|
| 540 |
+
"loss": 0.8604,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.0,
|
| 545 |
+
"eval_loss": 0.7618905305862427,
|
| 546 |
+
"eval_runtime": 129.5439,
|
| 547 |
+
"eval_samples_per_second": 39.407,
|
| 548 |
+
"eval_steps_per_second": 1.235,
|
| 549 |
+
"step": 76
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.0,
|
| 553 |
+
"step": 76,
|
| 554 |
+
"total_flos": 2.244031216091136e+16,
|
| 555 |
+
"train_loss": 2.2036053005017733,
|
| 556 |
+
"train_runtime": 949.5806,
|
| 557 |
+
"train_samples_per_second": 10.213,
|
| 558 |
+
"train_steps_per_second": 0.08
|
| 559 |
+
}
|
| 560 |
+
],
|
| 561 |
+
"logging_steps": 1,
|
| 562 |
+
"max_steps": 76,
|
| 563 |
+
"num_input_tokens_seen": 0,
|
| 564 |
+
"num_train_epochs": 1,
|
| 565 |
+
"save_steps": 500,
|
| 566 |
+
"stateful_callbacks": {
|
| 567 |
+
"TrainerControl": {
|
| 568 |
+
"args": {
|
| 569 |
+
"should_epoch_stop": false,
|
| 570 |
+
"should_evaluate": false,
|
| 571 |
+
"should_log": false,
|
| 572 |
+
"should_save": true,
|
| 573 |
+
"should_training_stop": true
|
| 574 |
+
},
|
| 575 |
+
"attributes": {}
|
| 576 |
+
}
|
| 577 |
+
},
|
| 578 |
+
"total_flos": 2.244031216091136e+16,
|
| 579 |
+
"train_batch_size": 8,
|
| 580 |
+
"trial_name": null,
|
| 581 |
+
"trial_params": null
|
| 582 |
+
}
|