Hugging Face's logo

adamo1139
/
yi-34b-200k-rawrr-dpo-1

Text Generation
Transformers
Safetensors
llama
finetune
fine-tune
text-generation-inference
Inference Endpoints
Model card Files Community
1
adamo1139 commited on
Commit
67d26fb
·
verified ·
1 Parent(s): 184e02e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -1
README.md CHANGED
@@ -9,8 +9,9 @@ datasets:
9
  - adamo1139/rawrr_v1
10
  ---
11
 
 
12
 
13
- This model is Yi-34B-200K fine-tuned using DPO on rawrr_v1 dataset using QLoRA at ctx 200. I then merged the adapter with base model.
14
  This model is akin to raw LLaMa 65B, it's not meant to follow instructions but instead should be useful as base for further fine-tuning.
15
 
16
  Rawrr_v1 dataset made it so that this model issue less refusals, especially for benign topics, and is moreso completion focused rather than instruct focused.
 
9
  - adamo1139/rawrr_v1
10
  ---
11
 
12
+ NEW STRONGER RAWRR FINETUNE COMING SOON!
13
 
14
+ This model is Yi-34B-200K fine-tuned using DPO on rawrr_v1 dataset using QLoRA at ctx 200, lora_r 4 and lora_alpha 8. I then merged the adapter with base model.
15
  This model is akin to raw LLaMa 65B, it's not meant to follow instructions but instead should be useful as base for further fine-tuning.
16
 
17
  Rawrr_v1 dataset made it so that this model issue less refusals, especially for benign topics, and is moreso completion focused rather than instruct focused.