sometimesanotion/Base-Chocolatine-2-14B-Instruct-v2.0b3

merge

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

Merge Details

Merge Method

This model was merged using the Passthrough merge method.

Models Merged

The following models were included in the merge:

• sometimesanotion/Base-Chocolatine-2-14B-Instruct-v2.0b3-slerp + sometimesanotion/LoRA-256-Base-Qwenvergence

Configuration

The following YAML configuration was used to produce this model:

name:                Base-Chocolatine-2-14B-Instruct-v2.0b3-slerp
merge_method:        slerp
base_model:          sometimesanotion/Base-Qwenvergence-dare_ties
tokenizer_source:    base
dtype:               float32
out_dtype:           bfloat16
parameters:
  t:                   1.00
slices:
  - sources:
    - { layer_range:   [  0,  2 ], model: sometimesanotion/Base-Qwenvergence-dare_ties }
    - { layer_range:   [  0,  2 ], model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3 }
    parameters:          { t: [ 0.00, 0.00 ] }
  - sources:
    - { layer_range:   [  2,  7 ], model: sometimesanotion/Base-Qwenvergence-dare_ties }
    - { layer_range:   [  2,  7 ], model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3 }
    parameters:          { t: [ 0.00, 0.50 ] }
  - sources:
    - { layer_range:   [  7, 12 ], model: sometimesanotion/Base-Qwenvergence-dare_ties }
    - { layer_range:   [  7, 12 ], model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3 }
    parameters:          { t: [ 0.50, 1.00 ] }
  - sources:
    - { layer_range:   [ 12, 48 ], model: sometimesanotion/Base-Qwenvergence-dare_ties }
    - { layer_range:   [ 12, 48 ], model: jpacifico/Chocolatine-2-14B-Instruct-v2.0b3 }
---
name:                Base-Chocolatine-2-14B-Instruct-v2.0b3
merge_method:        passthrough
dtype:               float32
out_dtype:           bfloat16
models:
  - model:           sometimesanotion/Base-Chocolatine-2-14B-Instruct-v2.0b3-slerp+sometimesanotion/LoRA-256-Base-Qwenvergence