Model save
Browse files- README.md +101 -0
- all_results.json +9 -0
- generation_config.json +10 -0
- train_results.json +9 -0
- trainer_state.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,101 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: llama3.2
|
| 4 |
+
base_model: meta-llama/Llama-3.2-3B
|
| 5 |
+
tags:
|
| 6 |
+
- trl
|
| 7 |
+
- sft
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: llama_3_2_3B_tulu3
|
| 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 |
+
# llama_3_2_3B_tulu3
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Llama-3.2-3B](https://huggingface.co/meta-llama/Llama-3.2-3B) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.6899
|
| 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: 2e-05
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 8
|
| 46 |
+
- gradient_accumulation_steps: 4
|
| 47 |
+
- total_train_batch_size: 128
|
| 48 |
+
- total_eval_batch_size: 32
|
| 49 |
+
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 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 |
+
| 0.7972 | 0.0275 | 200 | 0.8336 |
|
| 59 |
+
| 0.7579 | 0.0551 | 400 | 0.7996 |
|
| 60 |
+
| 0.8037 | 0.0826 | 600 | 0.7918 |
|
| 61 |
+
| 0.7333 | 0.1101 | 800 | 0.7879 |
|
| 62 |
+
| 0.7871 | 0.1376 | 1000 | 0.7818 |
|
| 63 |
+
| 0.8135 | 0.1652 | 1200 | 0.7736 |
|
| 64 |
+
| 0.7612 | 0.1927 | 1400 | 0.7699 |
|
| 65 |
+
| 0.7421 | 0.2202 | 1600 | 0.7643 |
|
| 66 |
+
| 0.7451 | 0.2478 | 1800 | 0.7595 |
|
| 67 |
+
| 0.7388 | 0.2753 | 2000 | 0.7556 |
|
| 68 |
+
| 0.7707 | 0.3028 | 2200 | 0.7523 |
|
| 69 |
+
| 0.7063 | 0.3303 | 2400 | 0.7481 |
|
| 70 |
+
| 0.8091 | 0.3579 | 2600 | 0.7440 |
|
| 71 |
+
| 0.764 | 0.3854 | 2800 | 0.7407 |
|
| 72 |
+
| 0.714 | 0.4129 | 3000 | 0.7370 |
|
| 73 |
+
| 0.6745 | 0.4405 | 3200 | 0.7339 |
|
| 74 |
+
| 0.6771 | 0.4680 | 3400 | 0.7295 |
|
| 75 |
+
| 0.7419 | 0.4955 | 3600 | 0.7257 |
|
| 76 |
+
| 0.71 | 0.5230 | 3800 | 0.7223 |
|
| 77 |
+
| 0.6362 | 0.5506 | 4000 | 0.7189 |
|
| 78 |
+
| 0.7616 | 0.5781 | 4200 | 0.7159 |
|
| 79 |
+
| 0.676 | 0.6056 | 4400 | 0.7126 |
|
| 80 |
+
| 0.6732 | 0.6332 | 4600 | 0.7094 |
|
| 81 |
+
| 0.7017 | 0.6607 | 4800 | 0.7067 |
|
| 82 |
+
| 0.6796 | 0.6882 | 5000 | 0.7038 |
|
| 83 |
+
| 0.7065 | 0.7157 | 5200 | 0.7012 |
|
| 84 |
+
| 0.6318 | 0.7433 | 5400 | 0.6987 |
|
| 85 |
+
| 0.639 | 0.7708 | 5600 | 0.6965 |
|
| 86 |
+
| 0.7078 | 0.7983 | 5800 | 0.6949 |
|
| 87 |
+
| 0.7029 | 0.8258 | 6000 | 0.6933 |
|
| 88 |
+
| 0.6977 | 0.8534 | 6200 | 0.6921 |
|
| 89 |
+
| 0.6803 | 0.8809 | 6400 | 0.6911 |
|
| 90 |
+
| 0.703 | 0.9084 | 6600 | 0.6905 |
|
| 91 |
+
| 0.6819 | 0.9360 | 6800 | 0.6901 |
|
| 92 |
+
| 0.6327 | 0.9635 | 7000 | 0.6899 |
|
| 93 |
+
| 0.6685 | 0.9910 | 7200 | 0.6899 |
|
| 94 |
+
|
| 95 |
+
|
| 96 |
+
### Framework versions
|
| 97 |
+
|
| 98 |
+
- Transformers 4.48.3
|
| 99 |
+
- Pytorch 2.2.2+cu121
|
| 100 |
+
- Datasets 2.18.0
|
| 101 |
+
- Tokenizers 0.21.0
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9999655896218299,
|
| 3 |
+
"total_flos": 1437402890797056.0,
|
| 4 |
+
"train_loss": 0.7255030241491708,
|
| 5 |
+
"train_runtime": 38613.5919,
|
| 6 |
+
"train_samples": 929949,
|
| 7 |
+
"train_samples_per_second": 24.083,
|
| 8 |
+
"train_steps_per_second": 0.188
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": 128001,
|
| 6 |
+
"pad_token_id": 128004,
|
| 7 |
+
"temperature": 0.6,
|
| 8 |
+
"top_p": 0.9,
|
| 9 |
+
"transformers_version": "4.48.3"
|
| 10 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9999655896218299,
|
| 3 |
+
"total_flos": 1437402890797056.0,
|
| 4 |
+
"train_loss": 0.7255030241491708,
|
| 5 |
+
"train_runtime": 38613.5919,
|
| 6 |
+
"train_samples": 929949,
|
| 7 |
+
"train_samples_per_second": 24.083,
|
| 8 |
+
"train_steps_per_second": 0.188
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|