metadata
base_model:
- Sela223/Repose-Marlin-12B
- Sela223/Captain-Foxfire-12B
library_name: transformers
tags:
- mergekit
- merge
merged
This is a merge of pre-trained language models created using mergekit.
Merge Details
Merge Method
This model was merged using the SLERP merge method.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
base_model: Sela223/Captain-Foxfire-12B
dtype: bfloat16
merge_method: slerp
tokenizer_source: base
slices:
- sources:
- model: Sela223/Captain-Foxfire-12B
layer_range: [0, 40]
- model: Sela223/Repose-Marlin-12B
layer_range: [0, 40]
parameters:
rescale: true
t:
- filter: ".*(q_proj|k_proj|v_proj).*"
value: [0.0, 0.1, 0.25, 0.4, 0.5, 0.5, 0.5, 0.5, 0.4, 0.25, 0.1, 0.0]
- filter: ".*o_proj.*"
value: [0.0, 0.1, 0.2, 0.35, 0.5, 0.5, 0.5, 0.5, 0.35, 0.2, 0.1, 0.0]
- filter: self_attn
value: [0.0, 0.1, 0.25, 0.4, 0.5, 0.5, 0.5, 0.5, 0.4, 0.25, 0.1, 0.0]
- filter: ".*(gate_proj|up_proj|down_proj).*"
value: [0.0, 0.15, 0.3, 0.45, 0.5, 0.5, 0.5, 0.5, 0.45, 0.3, 0.15, 0.0]
- filter: mlp
value: [0.0, 0.15, 0.3, 0.45, 0.5, 0.5, 0.5, 0.5, 0.45, 0.3, 0.15, 0.0]
- filter: ".*(input_layernorm|post_attention_layernorm|layernorm).*"
value: [0.0, 0.3, 0.5, 0.6, 0.4, 0.0, 0.0, 0.4, 0.6, 0.5, 0.3, 0.0]
- filter: "^(embed_tokens|lm_head)$"
value: 0.5
- value: [0.0, 0.3, 0.5, 0.6, 0.4, 0.0, 0.0, 0.4, 0.6, 0.5, 0.3, 0.0]