|
--- |
|
datasets: |
|
- jondurbin/airoboros-gpt4-1.4.1 |
|
--- |
|
# RoPE Scaled QLoRA Finetune of airoboros-33b-gpt4-1.4.1 (LoRA) |
|
|
|
Full model card with merged GPTQ 4bit quantized weights can be found here: https://huggingface.co/bhenrym14/airoboros-33b-gpt4-1.4.1-PI-8192-GPTQ |
|
|
|
fp16 merged weights can be found here: https://huggingface.co/bhenrym14/airoboros-33b-gpt4-1.4.1-PI-8192-fp16 |
|
|
|
## Overview |
|
|
|
This is [Jon Durbin's Airoboros 33B GPT4 1.4](https://huggingface.co/jondurbin/airoboros-33b-gpt4-1.4) (LoRA weights) with several key modifications: |
|
- Context length extended to 8192 by RoPE Scaled Embeddings, but NOT via the superHOT LoRA. I started with base Llama-33b. |
|
- Training sequences beyond 2048 have the target truncated to equal 2048. |
|
- Used airoboros-gpt4-1.4.1 dataset instead of airoboros-gpt4-1.4 |
|
|
|
Otherwise, I emulated the training process as closely as possible (rank 64 QLoRA) It was trained on 1x RTX 6000 Ada for ~43 hours. |
