---
license: apache-2.0
tags:
- CoT
model-index:
- name: falcon-1b-cot-t2
  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: 24.74
      name: normalized accuracy
    source:
      url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=aloobun/falcon-1b-cot-t2
      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: 24.75
      name: normalized accuracy
    source:
      url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=aloobun/falcon-1b-cot-t2
      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: 23.12
      name: accuracy
    source:
      url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=aloobun/falcon-1b-cot-t2
      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: 48.38
    source:
      url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=aloobun/falcon-1b-cot-t2
      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: 50.36
      name: accuracy
    source:
      url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=aloobun/falcon-1b-cot-t2
      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: 0.0
      name: accuracy
    source:
      url: https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard?query=aloobun/falcon-1b-cot-t2
      name: Open LLM Leaderboard
---

### Prompt template : chatml

fintuned for CoT reasoning.

```
from transformers import AutoTokenizer, AutoModelForCausalLM
import transformers
import torch

model = "aloobun/falcon-1b-cot-t2"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
sequences = pipeline(
   "<|im_start|>user\nDoes P=NP?<|im_end|>\n<|im_start|>assistant\n",
    max_length=256,
    do_sample=True,
    top_k=10,
    num_return_sequences=1,
    eos_token_id=tokenizer.eos_token_id,
)
for seq in sequences:
    print(f"Result: {seq['generated_text']}")
```

>Fine-tuning language models is like tuning the strings of an AI banjo in the cosmic saloon of the digital frontier. We're not just slinging code; it's a harmonious quest to shape the minds of silicon wanderers, crafting binary ballads and electronic echoes. Picture it as cybernetic bardic magic, where we, the tech sorcerers, weave algorithms with strands of imagination. But, in this cosmic hoedown, there's a twist – as we twang the strings of artificial intelligence, we're also seeding the algorithms with a bit of human stardust, adding quirks and quirksome biases. So, as we two-step into this frontier of creation, are we summoning AI troubadours of the future or just conjuring interstellar jesters, spinning tales of silicon whimsy and digital campfire banter?
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co./datasets/open-llm-leaderboard/details_aloobun__falcon-1b-cot-t2)

|             Metric              |Value|
|---------------------------------|----:|
|Avg.                             |28.56|
|AI2 Reasoning Challenge (25-Shot)|24.74|
|HellaSwag (10-Shot)              |24.75|
|MMLU (5-Shot)                    |23.12|
|TruthfulQA (0-shot)              |48.38|
|Winogrande (5-shot)              |50.36|
|GSM8k (5-shot)                   | 0.00|