README.md

metadata

license: apache-2.0
tags:
  - merge
  - mergekit
  - lazymergekit
  - NousResearch/Nous-Hermes-2-Yi-34B
  - jondurbin/bagel-dpo-34b-v0.2

HermesBagel-34B-v0.1

HermesBagel-34B-v0.1 is a merge of the following models using LazyMergekit:

• NousResearch/Nous-Hermes-2-Yi-34B
• jondurbin/bagel-dpo-34b-v0.2

🧩 Configuration

slices:
  - sources:
      - model: NousResearch/Nous-Hermes-2-Yi-34B
        layer_range: [0, 60]
      - model: jondurbin/bagel-dpo-34b-v0.2
        layer_range: [0, 60]
merge_method: slerp
base_model: NousResearch/Nous-Hermes-2-Yi-34B
parameters:
  t:
    - filter: self_attn
      value: [0, 0.5, 0.3, 0.7, 1]
    - filter: mlp
      value: [1, 0.5, 0.7, 0.3, 0]
    - value: 0.5
dtype: bfloat16

Basic Usage

Setup

!pip install -qU transformers accelerate

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model = "dfurman/HermesBagel-34B-v0.1"

tokenizer = AutoTokenizer.from_pretrained(model)

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

messages = [
    {"role": "user", "content": "What is a large language model?"},
]

print("\n\n*** Prompt:")
input_ids = tokenizer.apply_chat_template(
    messages,
    tokenize=True,
    return_tensors="pt",
)
print(tokenizer.decode(input_ids[0]))

print("\n\n*** Generate:")
with torch.autocast("cuda", dtype=torch.bfloat16):
    output = model.generate(
        input_ids=input_ids.to("cuda"),
        max_new_tokens=256,
        return_dict_in_generate=True,
        do_sample=True, 
        temperature=0.7, 
        top_k=50, 
        top_p=0.95
    )

response = tokenizer.decode(
    output["sequences"][0][len(input_ids[0]):], 
    skip_special_tokens=True
)
print(response)

Outputs

"""
coming
"""