README.md

---
library_name: transformers
license: apache-2.0
datasets:
- kaist-ai/CoT-Collection
tags:
- finetune
- gpt4
- synthetic data
- custom_code
- h2oai
---

![Cypher aloobun h2oai1.8B](https://i.imgur.com/2R6f4EX.jpeg)
- This is an experimental model, Finetuned [h2oai/h2o-danube-1.8b-chat](https://huggingface.co/h2oai/h2o-danube-1.8b-chat), on variety of CoT tasks.
- The original idea was to use this 1.8B model, divide the dataset based on task specific capabilities, train models and transform them into a mixture of experts.
- Hyperparameters: adamw with eps of 1e-8, cosine decay w/ 20% warmup, lr=2e-5.

## Format:
```
<|system|></s><|prompt|></s><|answer|>
```

## Benchamrks:

WIP

## Example:
```
from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer, StoppingCriteria
import torch

class MyStoppingCriteria(StoppingCriteria):
  def __init__(self, target_sequence, prompt):
    self.target_sequence = target_sequence
    self.prompt=prompt

  def __call__(self, input_ids, scores, **kwargs):
    generated_text = tokenizer.decode(input_ids[0])
    generated_text = generated_text.replace(self.prompt,'')
    if self.target_sequence in generated_text:
        return True 
    return False 

  def __len__(self):
    return 1

  def __iter__(self):
    yield self

modelpath="aloobun/Cypher-CoT-1.8B"

model = AutoModelForCausalLM.from_pretrained(
    modelpath,
    torch_dtype=torch.bfloat16,
    device_map="cuda",
    trust_remote_code=True,       
)

tokenizer = AutoTokenizer.from_pretrained(
    modelpath,
    trust_remote_code=True,      
    use_fast=False,
)

prompt = "<|prompt|>Avery opens a flower shop. She ties 8 bunches of flowers with 9 flowers in each bunch. How many bunches would she have if she put 12 flowers in each bunch instead?</s><|answer|>"
encoded_input = tokenizer(prompt, return_tensors='pt')
input_ids=encoded_input['input_ids'].cuda()
streamer = TextStreamer(tokenizer=tokenizer, skip_prompt=True)
op = model.generate(
    input_ids,
    streamer=streamer,
    pad_token_id=tokenizer.eos_token_id,
    do_sample=True,
    temperature=0.7,
    top_p=0.8,
    max_new_tokens=512,
    stopping_criteria=MyStoppingCriteria("</s>", prompt)
)
```

## Output: