convo_bot_gpt_v1 / README.md
Sharathhebbar24's picture
Adding Evaluation Results (#1)
364a5fd verified
metadata
language:
  - en
license: apache-2.0
datasets:
  - vicgalle/alpaca-gpt4
metrics:
  - perplexity
pipeline_tag: text-generation
model-index:
  - name: convo_bot_gpt_v1
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: AI2 Reasoning Challenge (25-Shot)
          type: ai2_arc
          config: ARC-Challenge
          split: test
          args:
            num_few_shot: 25
        metrics:
          - type: acc_norm
            value: 22.35
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Sharathhebbar24/convo_bot_gpt_v1
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: HellaSwag (10-Shot)
          type: hellaswag
          split: validation
          args:
            num_few_shot: 10
        metrics:
          - type: acc_norm
            value: 31.07
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Sharathhebbar24/convo_bot_gpt_v1
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU (5-Shot)
          type: cais/mmlu
          config: all
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 26.12
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Sharathhebbar24/convo_bot_gpt_v1
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: TruthfulQA (0-shot)
          type: truthful_qa
          config: multiple_choice
          split: validation
          args:
            num_few_shot: 0
        metrics:
          - type: mc2
            value: 38.71
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Sharathhebbar24/convo_bot_gpt_v1
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: Winogrande (5-shot)
          type: winogrande
          config: winogrande_xl
          split: validation
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 51.54
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Sharathhebbar24/convo_bot_gpt_v1
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GSM8k (5-shot)
          type: gsm8k
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 0
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=Sharathhebbar24/convo_bot_gpt_v1
          name: Open LLM Leaderboard

This model is a finetuned version of gpt2 using vicgalle/alpaca-gpt4

Model description

GPT-2 is a transformers model pre-trained on a very large corpus of English data in a self-supervised fashion. This means it was pre-trained on the raw texts only, with no humans labeling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely, it was trained to guess the next word in sentences.

More precisely, inputs are sequences of continuous text of a certain length and the targets are the same sequence, shifting one token (word or piece of word) to the right. The model uses a masking mechanism to make sure the predictions for the token i only use the inputs from 1 to i but not the future tokens.

This way, the model learns an inner representation of the English language that can then be used to extract features useful for downstream tasks. The model is best at what it was trained for, however, which is generating texts from a prompt.

To use this model

>>> from transformers import AutoTokenizer, AutoModelForCausalLM
>>> model_name = "Sharathhebbar24/convo_bot_gpt2"
>>> model = AutoModelForCausalLM.from_pretrained(model_name)
>>> tokenizer = AutoTokenizer.from_pretrained("gpt2")
>>> def generate_text(prompt):
>>>  inputs = tokenizer.encode(prompt, return_tensors='pt')
>>>  outputs = mod1.generate(inputs, max_length=64, pad_token_id=tokenizer.eos_token_id)
>>>  generated = tokenizer.decode(outputs[0], skip_special_tokens=True)
>>>  return generated
>>>  prompt = """
>>>  Below is an instruction that describes a task. Write a response that appropriately completes the request. 
>>>  ### Instruction: Who is the world's most famous painter? 
>>>  ###
>>>  """
>>> res = generate_text(prompt)
>>> res

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric Value
Avg. 28.30
AI2 Reasoning Challenge (25-Shot) 22.35
HellaSwag (10-Shot) 31.07
MMLU (5-Shot) 26.12
TruthfulQA (0-shot) 38.71
Winogrande (5-shot) 51.54
GSM8k (5-shot) 0.00