File size: 2,555 Bytes
9ec6e02
 
 
 
 
 
 
 
 
 
4c3ced6
 
be5506c
 
9ec6e02
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
be5506c
9ec6e02
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
---
license: llama3
library_name: peft
tags:
- axolotl
- generated_from_trainer
base_model: meta-llama/Meta-Llama-3-8B-Instruct
model-index:
- name: llama3-8b-instruct-summary
  results: []
language:
- en
datasets:
- ibivibiv/summary_instruct
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

[<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
<details><summary>See axolotl config</summary>

axolotl version: `0.4.0`
```yaml
adapter: qlora
base_model: meta-llama/Meta-Llama-3-8B-Instruct
base_model_config: meta-llama/Meta-Llama-3-8B-Instruct
datasets:
- path: ibivibiv/summary_instruct
  type: alpaca
flash_attention: true
gradient_accumulation_steps: 4
gradient_checkpointing: true
hf_use_auth_token: true
hub_model_id: ibivibiv/llama3-8b-instruct-summary
learning_rate: 0.0002
load_in_4bit: true
logging_steps: 1
lora_alpha: 16
lora_dropout: 0.05
lora_r: 32
lora_target_linear: true
lr_scheduler: cosine
micro_batch_size: 2
model_type: AutoModelForCausalLM
num_epochs: 3
optimizer: paged_adamw_32bit
output_dir: /job/out
sample_packing: true
save_safetensors: true
sequence_len: 4096
special_tokens:
  pad_token: <|end_of_text|>
tokenizer_type: AutoTokenizer
wandb_project: TuneStudio
wandb_run_id: summllamma
wandb_watch: 'true'
warmup_steps: 10

```

</details><br>

# llama3-8b-instruct-summary

This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the ibivibiv/summary_instruct dataset.

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0002
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 10
- num_epochs: 3

### Training results



### Framework versions

- PEFT 0.10.0
- Transformers 4.40.2
- Pytorch 2.1.2+cu118
- Datasets 2.19.1
- Tokenizers 0.19.1