README.md

metadata

base_model: Wonder-Griffin/Judge-GPT2
datasets:
  - stanfordnlp/imdb
language:
  - en
library_name: transformers
license: unlicense
pipeline_tag: text-generation
model-index:
  - name: Wonder-Griffin/JudgeLLM
    results:
      - task:
          type: text-classification
          name: Text Classification
          modality: nlp
        subtasks:
          - type: acceptability-classification
            name: Acceptability Classification
          - type: sentiment-analysis
            name: Sentiment Analysis
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.95
      - task:
          type: summarization
          name: Summarization
          modality: nlp
        metrics:
          - name: Rouge-L
            type: rouge
            value: 0.8
inference:
  parameters:
    aggregation_strategy: simple
    top_k: 10
    top_p: 0.9
    temperature: 0.7
    max_new_tokens: 50
    do_sample: true
    guidance_scale: 7.5
    num_inference_steps: 50
example_inputs:
  - text: Sample input text for text classification
    table:
      headers:
        - Column 1
        - Column 2
      rows:
        - - Data 1
          - Data 2
    src: path/to/asset
    prompt: Generate an image with the following prompt...
    candidate_labels:
      - positive
      - negative
    multi_class: true
    messages:
      - role: user
        content: What is the weather like today?
      - role: assistant
        content: The weather is sunny with a chance of rain.
tags:
  - text-generation-inference
  - question-answering

JudgeLLM

This model is a fine-tuned version of Wonder-Griffin/Judge-GPT2 on

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Framework versions

• Transformers 4.43.3
• Pytorch 2.4.0+cu124
• Datasets 2.20.0
• Tokenizers 0.19.1