README.md

---
language:
- en
license: other
tags:
- mlx
datasets:
- ehartford/dolphin
- jondurbin/airoboros-2.2.1
- ehartford/samantha-data
- ehartford/WizardLM_evol_instruct_V2_196k_unfiltered_merged_split
license_name: yi-license
license_link: LICENSE
model-index:
- name: dolphin-2.2-yi-34b-200k
  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: 42.15
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ehartford/dolphin-2.2-yi-34b-200k
      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: 68.18
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ehartford/dolphin-2.2-yi-34b-200k
      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: 55.47
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ehartford/dolphin-2.2-yi-34b-200k
      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: 45.93
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ehartford/dolphin-2.2-yi-34b-200k
      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: 64.56
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ehartford/dolphin-2.2-yi-34b-200k
      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: 3.71
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=ehartford/dolphin-2.2-yi-34b-200k
      name: Open LLM Leaderboard
---

# mlx-community/dolphin-2.2-yi-34b-200k-4bit
This model was converted to MLX format from [`cognitivecomputations/dolphin-2.2-yi-34b-200k`]() using mlx-lm version **0.7.0**.
Refer to the [original model card](https://huggingface.co/cognitivecomputations/dolphin-2.2-yi-34b-200k) for more details on the model.
## Use with mlx

```bash
pip install mlx-lm
```

```python
from mlx_lm import load, generate

model, tokenizer = load("mlx-community/dolphin-2.2-yi-34b-200k-4bit")
response = generate(model, tokenizer, prompt="hello", verbose=True)
```