File size: 1,300 Bytes
d6031e6 b4082e3 d6031e6 |
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 |
---
license: llama3
base_model: OwenArli/ArliAI-Llama-3-8B-Cumulus-v0.3
pipeline_tag: text-generation
tags:
- llama
- conversational
---
# QuantFactory/ArliAI-Llama-3-8B-Cumulus-v0.3-GGUF
This is quantized version of [OwenArli/ArliAI-Llama-3-8B-Cumulus-v0.3](https://huggingface.co./OwenArli/ArliAI-Llama-3-8B-Cumulus-v0.3) created using llama.cpp
# Model Description
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
In terms of reasoning and intelligence, this model is probably a bit worse than the OG model because of the decensoring. However, this model is better at long back and forth chats and will refuse less.
OpenLLM Benchmark:
Training:
- Full 8192 sequence length.
- Training duration is around 2 days on an RTX 4090, using 4-bit loading and Qlora 64-rank 64-alpha resulting in ~2% trainable weights.
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|>
``` |