RichardErkhov/chargoddard_-_mistral-11b-slimorca-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

mistral-11b-slimorca - GGUF

• Model creator: https://huggingface.co/chargoddard/
• Original model: https://huggingface.co/chargoddard/mistral-11b-slimorca/

Name	Quant method	Size
mistral-11b-slimorca.Q2_K.gguf	Q2_K	3.73GB
mistral-11b-slimorca.IQ3_XS.gguf	IQ3_XS	4.14GB
mistral-11b-slimorca.IQ3_S.gguf	IQ3_S	4.37GB
mistral-11b-slimorca.Q3_K_S.gguf	Q3_K_S	4.34GB
mistral-11b-slimorca.IQ3_M.gguf	IQ3_M	4.51GB
mistral-11b-slimorca.Q3_K.gguf	Q3_K	4.84GB
mistral-11b-slimorca.Q3_K_M.gguf	Q3_K_M	4.84GB
mistral-11b-slimorca.Q3_K_L.gguf	Q3_K_L	5.26GB
mistral-11b-slimorca.IQ4_XS.gguf	IQ4_XS	5.43GB
mistral-11b-slimorca.Q4_0.gguf	Q4_0	5.66GB
mistral-11b-slimorca.IQ4_NL.gguf	IQ4_NL	5.72GB
mistral-11b-slimorca.Q4_K_S.gguf	Q4_K_S	5.7GB
mistral-11b-slimorca.Q4_K.gguf	Q4_K	6.02GB
mistral-11b-slimorca.Q4_K_M.gguf	Q4_K_M	6.02GB
mistral-11b-slimorca.Q4_1.gguf	Q4_1	6.27GB
mistral-11b-slimorca.Q5_0.gguf	Q5_0	6.89GB
mistral-11b-slimorca.Q5_K_S.gguf	Q5_K_S	6.89GB
mistral-11b-slimorca.Q5_K.gguf	Q5_K	7.08GB
mistral-11b-slimorca.Q5_K_M.gguf	Q5_K_M	7.08GB
mistral-11b-slimorca.Q5_1.gguf	Q5_1	7.51GB
mistral-11b-slimorca.Q6_K.gguf	Q6_K	8.2GB
mistral-11b-slimorca.Q8_0.gguf	Q8_0	10.62GB

Original model description:

language: - en license: apache-2.0 datasets: - Open-Orca/SlimOrca base_model: mistralai/Mistral-7B-v0.1 model-index: - name: mistral-11b-slimorca 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: 64.25 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/mistral-11b-slimorca 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: 83.81 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/mistral-11b-slimorca 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: 63.66 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/mistral-11b-slimorca 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: 54.66 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/mistral-11b-slimorca 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: 77.98 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/mistral-11b-slimorca 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: 52.39 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=chargoddard/mistral-11b-slimorca name: Open LLM Leaderboard

Full weight fine tuned on two epochs of SlimOrca. Uses Mistral Instruct's prompt format.

The base model for this came from a variation on Undi's Mistral 11B recipe. The o_proj and down_proj tensors were set to zero in the added layers, making the output exactly identical to Mistral 7B before training.

Benchmarks look good locally but still evaluating actual usefulness. Update: this turned out great! 10/10 would recommend as a training approach.

Reproducing

This mergekit config was used to produce the base model:

slices:
  - sources:
      - model: mistralai/Mistral-7B-v0.1
        layer_range: [0, 24]
  - sources: # add middle layers with residuals scaled to zero
      - model: mistralai/Mistral-7B-v0.1
        layer_range: [8, 24]
        parameters:
          scale:
            - filter: o_proj
              value: 0.0
            - filter: down_proj
              value: 0.0
            - value: 1.0
  - sources:
      - model: mistralai/Mistral-7B-v0.1
        layer_range: [24, 32]
merge_method: passthrough
dtype: bfloat16

The axolotl config for fine tuning is available here.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	66.12
AI2 Reasoning Challenge (25-Shot)	64.25
HellaSwag (10-Shot)	83.81
MMLU (5-Shot)	63.66
TruthfulQA (0-shot)	54.66
Winogrande (5-shot)	77.98
GSM8k (5-shot)	52.39