You can deploy this model and start earning money today!

RichardErkhov/
vidhirambhia_-_mathcoder_8b_ties_0.2_weight-merged-gguf

Name: vidhirambhia_-_mathcoder_8b_ties_0.2_weight-merged-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

mathcoder_8b_ties_0.2_weight-merged - GGUF

Model creator: https://huggingface.co/vidhirambhia/
Original model: https://huggingface.co/vidhirambhia/mathcoder_8b_ties_0.2_weight-merged/

Name	Quant method	Size
mathcoder_8b_ties_0.2_weight-merged.Q2_K.gguf	Q2_K	2.96GB
mathcoder_8b_ties_0.2_weight-merged.IQ3_XS.gguf	IQ3_XS	3.28GB
mathcoder_8b_ties_0.2_weight-merged.IQ3_S.gguf	IQ3_S	3.43GB
mathcoder_8b_ties_0.2_weight-merged.Q3_K_S.gguf	Q3_K_S	3.41GB
mathcoder_8b_ties_0.2_weight-merged.IQ3_M.gguf	IQ3_M	3.52GB
mathcoder_8b_ties_0.2_weight-merged.Q3_K.gguf	Q3_K	3.74GB
mathcoder_8b_ties_0.2_weight-merged.Q3_K_M.gguf	Q3_K_M	3.74GB
mathcoder_8b_ties_0.2_weight-merged.Q3_K_L.gguf	Q3_K_L	4.03GB
mathcoder_8b_ties_0.2_weight-merged.IQ4_XS.gguf	IQ4_XS	4.18GB
mathcoder_8b_ties_0.2_weight-merged.Q4_0.gguf	Q4_0	4.34GB
mathcoder_8b_ties_0.2_weight-merged.IQ4_NL.gguf	IQ4_NL	4.38GB
mathcoder_8b_ties_0.2_weight-merged.Q4_K_S.gguf	Q4_K_S	4.37GB
mathcoder_8b_ties_0.2_weight-merged.Q4_K.gguf	Q4_K	4.58GB
mathcoder_8b_ties_0.2_weight-merged.Q4_K_M.gguf	Q4_K_M	4.58GB
mathcoder_8b_ties_0.2_weight-merged.Q4_1.gguf	Q4_1	4.78GB
mathcoder_8b_ties_0.2_weight-merged.Q5_0.gguf	Q5_0	5.21GB
mathcoder_8b_ties_0.2_weight-merged.Q5_K_S.gguf	Q5_K_S	5.21GB
mathcoder_8b_ties_0.2_weight-merged.Q5_K.gguf	Q5_K	5.34GB
mathcoder_8b_ties_0.2_weight-merged.Q5_K_M.gguf	Q5_K_M	5.34GB
mathcoder_8b_ties_0.2_weight-merged.Q5_1.gguf	Q5_1	5.65GB
mathcoder_8b_ties_0.2_weight-merged.Q6_K.gguf	Q6_K	6.14GB
mathcoder_8b_ties_0.2_weight-merged.Q8_0.gguf	Q8_0	7.95GB

Original model description:

base_model:

meta-llama/Llama-3.1-8B
MathGenie/MathCoder2-Llama-3-8B library_name: transformers tags:
mergekit
merge

mathcoder_8b_ties-merged

This is a merge of pre-trained language models created using mergekit.

Merge Details

Merge Method

This model was merged using the TIES merge method using meta-llama/Llama-3.1-8B as a base.

Models Merged

The following models were included in the merge:

MathGenie/MathCoder2-Llama-3-8B

Configuration

The following YAML configuration was used to produce this model:

yaml

base_model: meta-llama/Llama-3.1-8B
dtype: bfloat16
merge_method: ties
parameters:
  int8_mask: 1.0
  normalize: 1.0
slices:
- sources:
  - layer_range: [0, 32]
    model: meta-llama/Llama-3.1-8B
  - layer_range: [0, 32]
    model: MathGenie/MathCoder2-Llama-3-8B
    parameters:
      density:
      - value: 1.0
      weight:
      - value: 0.2
tokenizer_source: union

RichardErkhov/vidhirambhia_-_mathcoder_8b_ties_0.2_weight-merged-gguf