README.md

metadata

language:
  - en
license: llama3.2
tags:
  - text-generation-inference
  - transformers
  - llama
  - trl
  - sft
  - reasoning
  - llama-3
base_model: meta-llama/Llama-3.2-3B-Instruct
datasets:
  - KingNish/reasoning-base-20k
model-index:
  - name: thea-c-3b-25r
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: HuggingFaceH4/ifeval
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 74.02
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-c-3b-25r
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: BBH
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 22.77
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-c-3b-25r
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: hendrycks/competition_math
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 13.37
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-c-3b-25r
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 2.01
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-c-3b-25r
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MuSR (0-shot)
          type: TAUR-Lab/MuSR
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 1.27
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-c-3b-25r
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU-PRO (5-shot)
          type: TIGER-Lab/MMLU-Pro
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 24.2
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=piotr25691/thea-c-3b-25r
          name: Open LLM Leaderboard

Model Description

A work in progress reasoning Llama 3.2 3B model trained on reasoning data.

Since I used different training code, it is unknown whether it generates the same kind of reasoning. Here is what inference code you should use:

from transformers import AutoModelForCausalLM, AutoTokenizer

MAX_REASONING_TOKENS = 1024
MAX_RESPONSE_TOKENS = 512

model_name = "piotr25691/thea-3b-25r"

model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained(model_name)

prompt = "Which is greater 9.9 or 9.11 ??"
messages = [
    {"role": "user", "content": prompt}
]

# Generate reasoning
reasoning_template = tokenizer.apply_chat_template(messages, tokenize=False, add_reasoning_prompt=True)
reasoning_inputs = tokenizer(reasoning_template, return_tensors="pt").to(model.device)
reasoning_ids = model.generate(**reasoning_inputs, max_new_tokens=MAX_REASONING_TOKENS)
reasoning_output = tokenizer.decode(reasoning_ids[0, reasoning_inputs.input_ids.shape[1]:], skip_special_tokens=True)

# print("REASONING: " + reasoning_output)

# Generate answer
messages.append({"role": "reasoning", "content": reasoning_output})
response_template = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
response_inputs = tokenizer(response_template, return_tensors="pt").to(model.device)
response_ids = model.generate(**response_inputs, max_new_tokens=MAX_RESPONSE_TOKENS)
response_output = tokenizer.decode(response_ids[0, response_inputs.input_ids.shape[1]:], skip_special_tokens=True)

print("ANSWER: " + response_output)

• Trained by: Piotr Zalewski
• License: llama3.2
• Finetuned from model: meta-llama/Llama-3.2-3B-Instruct
• Dataset used: KingNish/reasoning-base-20k

This Llama model was trained faster than Unsloth using custom training code.

Visit https://www.kaggle.com/code/piotr25691/distributed-llama-training-with-2xt4 to find out how you can finetune your models using BOTH of the Kaggle provided GPUs.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	22.94
IFEval (0-Shot)	74.02
BBH (3-Shot)	22.77
MATH Lvl 5 (4-Shot)	13.37
GPQA (0-shot)	2.01
MuSR (0-shot)	1.27
MMLU-PRO (5-shot)	24.20