metadata

license: apache-2.0
library_name: transformers
tags:
  - general-purpose
  - roleplay
  - storywriting
  - merge
  - finetune
base_model: mistralai/Mistral-Nemo-Base-2407
model-index:
  - name: Chronos-Gold-12B-1.0
    results:
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: IFEval (0-Shot)
          type: HuggingFaceH4/ifeval
          args:
            num_few_shot: 0
        metrics:
          - type: inst_level_strict_acc and prompt_level_strict_acc
            value: 31.66
            name: strict accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=elinas/Chronos-Gold-12B-1.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: BBH (3-Shot)
          type: BBH
          args:
            num_few_shot: 3
        metrics:
          - type: acc_norm
            value: 35.91
            name: normalized accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=elinas/Chronos-Gold-12B-1.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MATH Lvl 5 (4-Shot)
          type: hendrycks/competition_math
          args:
            num_few_shot: 4
        metrics:
          - type: exact_match
            value: 4.38
            name: exact match
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=elinas/Chronos-Gold-12B-1.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: GPQA (0-shot)
          type: Idavidrein/gpqa
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 9.06
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=elinas/Chronos-Gold-12B-1.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MuSR (0-shot)
          type: TAUR-Lab/MuSR
          args:
            num_few_shot: 0
        metrics:
          - type: acc_norm
            value: 19.42
            name: acc_norm
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=elinas/Chronos-Gold-12B-1.0
          name: Open LLM Leaderboard
      - task:
          type: text-generation
          name: Text Generation
        dataset:
          name: MMLU-PRO (5-shot)
          type: TIGER-Lab/MMLU-Pro
          config: main
          split: test
          args:
            num_few_shot: 5
        metrics:
          - type: acc
            value: 27.98
            name: accuracy
        source:
          url: >-
            https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=elinas/Chronos-Gold-12B-1.0
          name: Open LLM Leaderboard

Chronos Gold 12B-1.0

Chronos Gold 12B 1.0 is a very unique model that applies to domain areas such as general chatbot functionatliy, roleplay, and storywriting. The model has been observed to write up to 2250 tokens in a single sequence. The model was trained at a sequence length of 16384 (16k) and will still retain the apparent 128k context length from Mistral-Nemo, though it deteriorates over time like regular Nemo does based on the RULER Test

As a result, is recommended to keep your sequence length max at 16384, or you will experience performance degredation.

The base model is mistralai/Mistral-Nemo-Base-2407 which was heavily modified to produce a more coherent model, comparable to much larger models.

Chronos Gold 12B-1.0 re-creates the uniqueness of the original Chronos with significiantly enhanced prompt adherence (following), coherence, a modern dataset, as well as supporting a majority of "character card" formats in applications like SillyTavern.

It went through an iterative and objective merge process as my previous models and was further finetuned on a dataset curated for it.

The specifics of the model will not be disclosed at the time due to dataset ownership.

Instruct Template

This model uses ChatML - below is an example. It is a preset in many frontends.

<|im_start|>system
A system prompt describing how you'd like your bot to act.<|im_end|>
<|im_start|>user
Hello there!<|im_end|>
<|im_start|>assistant
I can assist you or we can discuss other things?<|im_end|>
<|im_start|>user
I was wondering how transformers work?<|im_end|>
<|im_start|>assistant

Quantization

Sampling Settings

Nemo is a bit sensitive to high temperatures, so I use lower. Here are my settings:

Temp - 0.7 (0.9 max)
Presence Penalty - 1.0
Repetition Penalty range - 2800
Min P - 0.10

Additional Details

This model was created by elinas on discord. Thank you to @kalomaze for providing a model that made this merge possible.

This is one of multiple models to come out in the series by size and model architecture, so look forward to it!

Contact me on Discord for inquiries.

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	21.40
IFEval (0-Shot)	31.66
BBH (3-Shot)	35.91
MATH Lvl 5 (4-Shot)	4.38
GPQA (0-shot)	9.06
MuSR (0-shot)	19.42
MMLU-PRO (5-shot)	27.98