base_model:
- akjindal53244/Llama-3.1-Storm-8B
- Sao10K/L3.1-8B-Niitama-v1.1
- v000000/L3.1-Niitorm-8B-t0.0001
library_name: transformers
tags:
- merge
- llama
- dpo
datasets:
- jondurbin/gutenberg-dpo-v0.1
Llama-3.1-Niitorm-8B-DPO
- DPO Trained, Llama3.1-8B.
New: DPO'd Gutenberg Version (full epoch training).
RP model, Niitama 1.1 as a base, nearswapped with one of the smartest 3.1 models "Storm", mostly abliterated.
Gutenberg dataset creates more human writer-like prose and greately lessen synthetic feeling outputs.
Finetune and merge
This is a merge and finetune of pre-trained language models.
Resultant merge finetuned on jondurbin/gutenberg-dpo-v0.1 for 1 epoch, 1.5e-5 learning rate, on Nvidia A100.
Merge Details
Merge Method
This model was merged using the NEARSWAP t0.0001 merge algorithm.
Models Merged
The following models were included in the merge:
- Base Model: Sao10K/L3.1-8B-Niitama-v1.1 + grimjim/Llama-3-Instruct-abliteration-LoRA-8B
- akjindal53244/Llama-3.1-Storm-8B
Configuration
The following YAML configuration was used to produce this model:
slices:
- sources:
- model: Sao10K/L3.1-8B-Niitama-v1.1+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
layer_range: [0, 32]
- model: akjindal53244/Llama-3.1-Storm-8B
layer_range: [0, 32]
merge_method: nearswap
base_model: Sao10K/L3.1-8B-Niitama-v1.1+grimjim/Llama-3-Instruct-abliteration-LoRA-8B
parameters:
t:
- value: 0.0001
dtype: bfloat16
# Then, DPO Finetune
Resultant merge finetuned on jondurbin/gutenberg-dpo-v0.1 for 1 epoch, 1.5e-5 learning rate, on Nvidia A100.
I used a higher learning rate and full dataset compared to "L3.1-Celestial-Stone-2x8B-DPO".
Prompt Template:
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
{system_prompt}<|eot_id|><|start_header_id|>user<|end_header_id|>
{input}<|eot_id|><|start_header_id|>assistant<|end_header_id|>
{output}<|eot_id|>
Credit to Alchemonaut.
Credit to jondurbin.
Credit to woofwolfy.