File size: 1,122 Bytes
843f94d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: llama3
---
Based on Meta-Llama-3-8b-Instruct, and is governed by Meta Llama 3 License agreement:
https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct


Experimental model trying to make Llama 3 8B better in general overall. Pretty difficult to do.


Base model: https://huggingface.co/failspy/Meta-Llama-3-8B-Instruct-abliterated-v3

Training:
- 4096 sequence length
- Training duration is around 2 days on 2x3090Ti
- 1 epoch training with a massive dataset for minimized repetition sickness.
- LORA with 64-rank 128-alpha resulting in ~2% trainable weights.


Llama 3 Instruct format:
```
<|begin_of_text|><|start_header_id|>system<|end_header_id|>

{{ system_prompt }}<|eot_id|><|start_header_id|>user<|end_header_id|>

{{ user_message_1 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>

{{ model_answer_1 }}<|eot_id|><|start_header_id|>user<|end_header_id|>

{{ user_message_2 }}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
```


Quants:

FP16: https://huggingface.co/OwenArli/ArliAI-Llama-3-8B-Argon-v1.0

GGUF: https://huggingface.co/OwenArli/ArliAI-Llama-3-8B-Argon-v1.0-GGUF