You can deploy this model and start earning money today!

RichardErkhov/
grimjim_-_llama-3-Nephilim-v3-8B-gguf

Name: grimjim_-_llama-3-Nephilim-v3-8B-gguf
Availability: InStock

Views

No views yet

Quantization made by Richard Erkhov.

Github

Discord

Request more models

llama-3-Nephilim-v3-8B - GGUF

Model creator: https://huggingface.co/grimjim/
Original model: https://huggingface.co/grimjim/llama-3-Nephilim-v3-8B/

Name	Quant method	Size
llama-3-Nephilim-v3-8B.Q2_K.gguf	Q2_K	2.96GB
llama-3-Nephilim-v3-8B.IQ3_XS.gguf	IQ3_XS	3.28GB
llama-3-Nephilim-v3-8B.IQ3_S.gguf	IQ3_S	3.43GB
llama-3-Nephilim-v3-8B.Q3_K_S.gguf	Q3_K_S	3.41GB
llama-3-Nephilim-v3-8B.IQ3_M.gguf	IQ3_M	3.52GB
llama-3-Nephilim-v3-8B.Q3_K.gguf	Q3_K	3.74GB
llama-3-Nephilim-v3-8B.Q3_K_M.gguf	Q3_K_M	3.74GB
llama-3-Nephilim-v3-8B.Q3_K_L.gguf	Q3_K_L	4.03GB
llama-3-Nephilim-v3-8B.IQ4_XS.gguf	IQ4_XS	4.18GB
llama-3-Nephilim-v3-8B.Q4_0.gguf	Q4_0	4.34GB
llama-3-Nephilim-v3-8B.IQ4_NL.gguf	IQ4_NL	4.38GB
llama-3-Nephilim-v3-8B.Q4_K_S.gguf	Q4_K_S	4.37GB
llama-3-Nephilim-v3-8B.Q4_K.gguf	Q4_K	4.58GB
llama-3-Nephilim-v3-8B.Q4_K_M.gguf	Q4_K_M	4.58GB
llama-3-Nephilim-v3-8B.Q4_1.gguf	Q4_1	4.78GB
llama-3-Nephilim-v3-8B.Q5_0.gguf	Q5_0	5.21GB
llama-3-Nephilim-v3-8B.Q5_K_S.gguf	Q5_K_S	5.21GB
llama-3-Nephilim-v3-8B.Q5_K.gguf	Q5_K	5.34GB
llama-3-Nephilim-v3-8B.Q5_K_M.gguf	Q5_K_M	5.34GB
llama-3-Nephilim-v3-8B.Q5_1.gguf	Q5_1	5.65GB
llama-3-Nephilim-v3-8B.Q6_K.gguf	Q6_K	6.14GB
llama-3-Nephilim-v3-8B.Q8_0.gguf	Q8_0	7.95GB

Original model description:

base_model:

grimjim/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge
tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1 library_name: transformers tags:
mergekit
merge license: cc-by-nc-4.0 pipeline_tag: text-generation

llama-3-Nephilim-v3-8B

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

GGUF quants are here.

Although none of the components of this merge were trained for roleplay nor intended for it, the model can be used effectively in that role.

Tested with temperature 1 and minP 0.01. This model leans toward being creative, so adjust temperature upward or downward as desired.

There are initial format consistency issues with the merged model, but this can be mitigated in an Instruct prompt. Additionally, promptsteering was employed to vary the text generation output to avoid some of the common failings observed during text generation with Llama 3 8B models. The complete Instruct prompt used during testing is available below.

Built with Meta Llama 3.

Merge Details

Merge Method

This model was merged using the task arithmetic merge method using grimjim/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge as a base.

Models Merged

The following models were included in the merge:

tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1

Configuration

The following YAML configuration was used to produce this model:

yaml

base_model: grimjim/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge
dtype: bfloat16
merge_method: task_arithmetic
parameters:
  normalize: false
slices:
- sources:
  - layer_range: [0, 32]
    model: grimjim/Llama-3-Instruct-8B-SPPO-Iter3-SimPO-merge
  - layer_range: [0, 32]
    model: tokyotech-llm/Llama-3-Swallow-8B-Instruct-v0.1
    parameters:
      weight: 0.1

RichardErkhov/grimjim_-_llama-3-Nephilim-v3-8B-gguf