You can deploy this model and start earning money today!

RichardErkhov/
EryriLabs_-_Llama-3.2-SARAThinker-merged-3b-gguf

Name: EryriLabs_-_Llama-3.2-SARAThinker-merged-3b-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

Llama-3.2-SARAThinker-merged-3b - GGUF

Model creator: https://huggingface.co/EryriLabs/
Original model: https://huggingface.co/EryriLabs/Llama-3.2-SARAThinker-merged-3b/

Name	Quant method	Size
Llama-3.2-SARAThinker-merged-3b.Q2_K.gguf	Q2_K	1.39GB
Llama-3.2-SARAThinker-merged-3b.IQ3_XS.gguf	IQ3_XS	1.53GB
Llama-3.2-SARAThinker-merged-3b.IQ3_S.gguf	IQ3_S	1.59GB
Llama-3.2-SARAThinker-merged-3b.Q3_K_S.gguf	Q3_K_S	1.59GB
Llama-3.2-SARAThinker-merged-3b.IQ3_M.gguf	IQ3_M	1.65GB
Llama-3.2-SARAThinker-merged-3b.Q3_K.gguf	Q3_K	1.73GB
Llama-3.2-SARAThinker-merged-3b.Q3_K_M.gguf	Q3_K_M	1.73GB
Llama-3.2-SARAThinker-merged-3b.Q3_K_L.gguf	Q3_K_L	1.85GB
Llama-3.2-SARAThinker-merged-3b.IQ4_XS.gguf	IQ4_XS	1.91GB
Llama-3.2-SARAThinker-merged-3b.Q4_0.gguf	Q4_0	1.99GB
Llama-3.2-SARAThinker-merged-3b.IQ4_NL.gguf	IQ4_NL	2.0GB
Llama-3.2-SARAThinker-merged-3b.Q4_K_S.gguf	Q4_K_S	2.0GB
Llama-3.2-SARAThinker-merged-3b.Q4_K.gguf	Q4_K	2.09GB
Llama-3.2-SARAThinker-merged-3b.Q4_K_M.gguf	Q4_K_M	2.09GB
Llama-3.2-SARAThinker-merged-3b.Q4_1.gguf	Q4_1	2.18GB
Llama-3.2-SARAThinker-merged-3b.Q5_0.gguf	Q5_0	2.37GB
Llama-3.2-SARAThinker-merged-3b.Q5_K_S.gguf	Q5_K_S	2.37GB
Llama-3.2-SARAThinker-merged-3b.Q5_K.gguf	Q5_K	2.41GB
Llama-3.2-SARAThinker-merged-3b.Q5_K_M.gguf	Q5_K_M	2.41GB
Llama-3.2-SARAThinker-merged-3b.Q5_1.gguf	Q5_1	2.55GB
Llama-3.2-SARAThinker-merged-3b.Q6_K.gguf	Q6_K	2.76GB
Llama-3.2-SARAThinker-merged-3b.Q8_0.gguf	Q8_0	3.58GB

Original model description:

base_model:

HollowMan6/Llama-3.2-3B-SFT-Model-Ocra-500k
Lyte/Llama-3.2-3B-Overthinker
EryriLabs/Llama-3.2-SARA-3b library_name: transformers tags:
mergekit
merge

Llama-3.2-SARAThinker-merged-3b

This is a merge of pre-trained language models created using mergekit. It was developed as part of an ongoing blog series. I take no responsibility for the ouputs of this model.

Merge Details

Merge Method

This model was merged using the TIES merge method using EryriLabs/Llama-3.2-SARA-3b as a base.

Models Merged

The following models were included in the merge:

Configuration

The following YAML configuration was used to produce this model:

yaml

models:
  - model: EryriLabs/Llama-3.2-SARA-3b
    parameters:
      density: 0.99  # keeping 90% of base model weights
  - model: Lyte/Llama-3.2-3B-Overthinker
    parameters:
      density: 0.1  # fraction of weights in differences from the base model to retain
      weight:   # weight gradient
        - filter: mlp
          value: 0.1
        - value: 0
  - model: HollowMan6/Llama-3.2-3B-SFT-Model-Ocra-500k
    parameters:
      density: 0.1
      weight: 0.2
merge_method: ties
base_model: EryriLabs/Llama-3.2-SARA-3b
parameters:
  normalize: true
  int8_mask: true
dtype: float16

RichardErkhov/EryriLabs_-_Llama-3.2-SARAThinker-merged-3b-gguf