TheSkullery
/

MD-Zephyria-42b

Text Generation

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

Edit model card

Model Information

Base Model: unsloth/Mistral-Small-Instruct-2409

Strategy: Mid Duplication

Total Layers: 55

Duplication Start: Layer 22 (40% of model)

Duplicated Layers: 27 (49.1% of model)

Unique Final Layers: 7 (12.7% of model)

Model Characteristics

Models down_proj and o_proj layers have been nulled and will require healing
Balances early feature extraction and later refinement
Even split between unique and duplicated sections
Good for general-purpose tasks with balanced low and high-level processing
May provide a good compromise for a wide range of applications

Configuration Visualization


[     Unique     ][     Duplicated     ][  Unique  ]
0 ------------- 21 22 ------------- 48 49 ------- 54
      40%              49.1%            10.9%

Downloads last month: 13

Safetensors

Model size

42.1B params

Tensor type

BF16

·

Inference Examples

Text Generation

This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for TheSkullery/MD-Zephyria-42b

Base model

mistralai/Mistral-Small-Instruct-2409

Finetuned

unsloth/Mistral-Small-Instruct-2409

Finetuned

(8)

this model

Quantizations