metadata
library_name: peft
tags:
- meta-llama
- code
- instruct
- WizardLM
- Mistral-7B-v0.1
datasets:
- WizardLM/WizardLM_evol_instruct_70k
base_model: HuggingFaceH4/zephyr-7b-alpha
license: apache-2.0
Finetuning Overview:
Model Used: HuggingFaceH4/zephyr-7b-alpha
Dataset: WizardLM/WizardLM_evol_instruct_70k
Dataset Insights:
The WizardLM/WizardLM_evol_instruct_70k dataset, tailored specifically for enhancing interactive capabilities, was developed using the EVOL-Instruct method. This method enhances a smaller dataset with tougher questions for the LLM to perform.
Finetuning Details:
With the utilization of MonsterAPI's LLM finetuner, this finetuning:
- Was achieved with great cost-effectiveness.
- Completed in a total duration of 5hrs 18mins for 1 epoch using an A6000 48GB GPU.
- Costed
$10
for the entire epoch.
Hyperparameters & Additional Details:
- Epochs: 1
- Cost Per Epoch: $10.5
- Total Finetuning Cost: $10.5
- Model Path: HuggingFaceH4/zephyr-7b-alpha
- Learning Rate: 0.0002
- Data Split: 90% train 10% validation
- Gradient Accumulation Steps: 4
Prompt Structure
### INSTRUCTION:
[instruction]
### RESPONSE:
[output]
Benchmarking results
ARC (arc_challenge, acc_norm) 0.5543 HellaSwag (hellaswag, acc_norm) 0.7979 TruthfulQA (truthfulqa_mc2) 0.4781
license: apache-2.0