YAML Metadata
Warning:
empty or missing yaml metadata in repo card
(https://huggingface.co/docs/hub/model-cards#model-card-metadata)
Merge of every weight using passthrough.
Thanks to chargoddard for this: https://huggingface.co/chargoddard/demixtral
Heres the mergekit yaml:
slices:
- sources:
- model: mixtral-expert-0
layer_range:
- 0
- 1
- sources:
- model: mixtral-expert-1
layer_range:
- 0
- 1
- sources:
- model: mixtral-expert-2
layer_range:
- 0
- 1
- sources:
- model: mixtral-expert-3
layer_range:
- 0
- 1
- sources:
- model: mixtral-expert-4
layer_range:
- 0
- 1
- sources:
- model: mixtral-expert-5
layer_range:
- 0
- 1
- sources:
- model: mixtral-expert-6
layer_range:
- 0
- 1
- sources:
- model: mixtral-expert-7
layer_range:
- 0
- 1
- sources:
- model: mixtral-expert-0
layer_range:
- 1
- 2
- sources:
- model: mixtral-expert-1
layer_range:
- 1
- 2
- sources:
- model: mixtral-expert-2
layer_range:
- 1
- 2
- sources:
- model: mixtral-expert-3
layer_range:
- 1
- 2
- sources:
- model: mixtral-expert-4
layer_range:
- 1
- 2
- sources:
- model: mixtral-expert-5
layer_range:
- 1
- 2
- sources:
- model: mixtral-expert-6
layer_range:
- 1
- 2
- sources:
- model: mixtral-expert-7
layer_range:
- 1
- 2
- sources:
- model: mixtral-expert-0
layer_range:
- 2
- 3
- sources:
- model: mixtral-expert-1
layer_range:
- 2
- 3
- sources:
- model: mixtral-expert-2
layer_range:
- 2
- 3
- sources:
- model: mixtral-expert-3
layer_range:
- 2
- 3
- sources:
- model: mixtral-expert-4
layer_range:
- 2
- 3
- sources:
- model: mixtral-expert-5
layer_range:
- 2
- 3
- sources:
- model: mixtral-expert-6
layer_range:
- 2
- 3
- sources:
- model: mixtral-expert-7
layer_range:
- 2
- 3
- sources:
- model: mixtral-expert-0
layer_range:
- 3
- 4
- sources:
- model: mixtral-expert-1
layer_range:
- 3
- 4
- sources:
- model: mixtral-expert-2
layer_range:
- 3
- 4
- sources:
- model: mixtral-expert-3
layer_range:
- 3
- 4
- sources:
- model: mixtral-expert-4
layer_range:
- 3
- 4
- sources:
- model: mixtral-expert-5
layer_range:
- 3
- 4
- sources:
- model: mixtral-expert-6
layer_range:
- 3
- 4
- sources:
- model: mixtral-expert-7
layer_range:
- 3
- 4
- sources:
- model: mixtral-expert-0
layer_range:
- 4
- 5
- sources:
- model: mixtral-expert-1
layer_range:
- 4
- 5
- sources:
- model: mixtral-expert-2
layer_range:
- 4
- 5
- sources:
- model: mixtral-expert-3
layer_range:
- 4
- 5
- sources:
- model: mixtral-expert-4
layer_range:
- 4
- 5
- sources:
- model: mixtral-expert-5
layer_range:
- 4
- 5
- sources:
- model: mixtral-expert-6
layer_range:
- 4
- 5
- sources:
- model: mixtral-expert-7
layer_range:
- 4
- 5
- sources:
- model: mixtral-expert-0
layer_range:
- 5
- 6
- sources:
- model: mixtral-expert-1
layer_range:
- 5
- 6
- sources:
- model: mixtral-expert-2
layer_range:
- 5
- 6
- sources:
- model: mixtral-expert-3
layer_range:
- 5
- 6
- sources:
- model: mixtral-expert-4
layer_range:
- 5
- 6
- sources:
- model: mixtral-expert-5
layer_range:
- 5
- 6
- sources:
- model: mixtral-expert-6
layer_range:
- 5
- 6
- sources:
- model: mixtral-expert-7
layer_range:
- 5
- 6
- sources:
- model: mixtral-expert-0
layer_range:
- 6
- 7
- sources:
- model: mixtral-expert-1
layer_range:
- 6
- 7
- sources:
- model: mixtral-expert-2
layer_range:
- 6
- 7
- sources:
- model: mixtral-expert-3
layer_range:
- 6
- 7
- sources:
- model: mixtral-expert-4
layer_range:
- 6
- 7
- sources:
- model: mixtral-expert-5
layer_range:
- 6
- 7
- sources:
- model: mixtral-expert-6
layer_range:
- 6
- 7
- sources:
- model: mixtral-expert-7
layer_range:
- 6
- 7
- sources:
- model: mixtral-expert-0
layer_range:
- 7
- 8
- sources:
- model: mixtral-expert-1
layer_range:
- 7
- 8
- sources:
- model: mixtral-expert-2
layer_range:
- 7
- 8
- sources:
- model: mixtral-expert-3
layer_range:
- 7
- 8
- sources:
- model: mixtral-expert-4
layer_range:
- 7
- 8
- sources:
- model: mixtral-expert-5
layer_range:
- 7
- 8
- sources:
- model: mixtral-expert-6
layer_range:
- 7
- 8
- sources:
- model: mixtral-expert-7
layer_range:
- 7
- 8
- sources:
- model: mixtral-expert-0
layer_range:
- 8
- 9
- sources:
- model: mixtral-expert-1
layer_range:
- 8
- 9
- sources:
- model: mixtral-expert-2
layer_range:
- 8
- 9
- sources:
- model: mixtral-expert-3
layer_range:
- 8
- 9
- sources:
- model: mixtral-expert-4
layer_range:
- 8
- 9
- sources:
- model: mixtral-expert-5
layer_range:
- 8
- 9
- sources:
- model: mixtral-expert-6
layer_range:
- 8
- 9
- sources:
- model: mixtral-expert-7
layer_range:
- 8
- 9
- sources:
- model: mixtral-expert-0
layer_range:
- 9
- 10
- sources:
- model: mixtral-expert-1
layer_range:
- 9
- 10
- sources:
- model: mixtral-expert-2
layer_range:
- 9
- 10
- sources:
- model: mixtral-expert-3
layer_range:
- 9
- 10
- sources:
- model: mixtral-expert-4
layer_range:
- 9
- 10
- sources:
- model: mixtral-expert-5
layer_range:
- 9
- 10
- sources:
- model: mixtral-expert-6
layer_range:
- 9
- 10
- sources:
- model: mixtral-expert-7
layer_range:
- 9
- 10
- sources:
- model: mixtral-expert-0
layer_range:
- 10
- 11
- sources:
- model: mixtral-expert-1
layer_range:
- 10
- 11
- sources:
- model: mixtral-expert-2
layer_range:
- 10
- 11
- sources:
- model: mixtral-expert-3
layer_range:
- 10
- 11
- sources:
- model: mixtral-expert-4
layer_range:
- 10
- 11
- sources:
- model: mixtral-expert-5
layer_range:
- 10
- 11
- sources:
- model: mixtral-expert-6
layer_range:
- 10
- 11
- sources:
- model: mixtral-expert-7
layer_range:
- 10
- 11
- sources:
- model: mixtral-expert-0
layer_range:
- 11
- 12
- sources:
- model: mixtral-expert-1
layer_range:
- 11
- 12
- sources:
- model: mixtral-expert-2
layer_range:
- 11
- 12
- sources:
- model: mixtral-expert-3
layer_range:
- 11
- 12
- sources:
- model: mixtral-expert-4
layer_range:
- 11
- 12
- sources:
- model: mixtral-expert-5
layer_range:
- 11
- 12
- sources:
- model: mixtral-expert-6
layer_range:
- 11
- 12
- sources:
- model: mixtral-expert-7
layer_range:
- 11
- 12
- sources:
- model: mixtral-expert-0
layer_range:
- 12
- 13
- sources:
- model: mixtral-expert-1
layer_range:
- 12
- 13
- sources:
- model: mixtral-expert-2
layer_range:
- 12
- 13
- sources:
- model: mixtral-expert-3
layer_range:
- 12
- 13
- sources:
- model: mixtral-expert-4
layer_range:
- 12
- 13
- sources:
- model: mixtral-expert-5
layer_range:
- 12
- 13
- sources:
- model: mixtral-expert-6
layer_range:
- 12
- 13
- sources:
- model: mixtral-expert-7
layer_range:
- 12
- 13
- sources:
- model: mixtral-expert-0
layer_range:
- 13
- 14
- sources:
- model: mixtral-expert-1
layer_range:
- 13
- 14
- sources:
- model: mixtral-expert-2
layer_range:
- 13
- 14
- sources:
- model: mixtral-expert-3
layer_range:
- 13
- 14
- sources:
- model: mixtral-expert-4
layer_range:
- 13
- 14
- sources:
- model: mixtral-expert-5
layer_range:
- 13
- 14
- sources:
- model: mixtral-expert-6
layer_range:
- 13
- 14
- sources:
- model: mixtral-expert-7
layer_range:
- 13
- 14
- sources:
- model: mixtral-expert-0
layer_range:
- 14
- 15
- sources:
- model: mixtral-expert-1
layer_range:
- 14
- 15
- sources:
- model: mixtral-expert-2
layer_range:
- 14
- 15
- sources:
- model: mixtral-expert-3
layer_range:
- 14
- 15
- sources:
- model: mixtral-expert-4
layer_range:
- 14
- 15
- sources:
- model: mixtral-expert-5
layer_range:
- 14
- 15
- sources:
- model: mixtral-expert-6
layer_range:
- 14
- 15
- sources:
- model: mixtral-expert-7
layer_range:
- 14
- 15
- sources:
- model: mixtral-expert-0
layer_range:
- 15
- 16
- sources:
- model: mixtral-expert-1
layer_range:
- 15
- 16
- sources:
- model: mixtral-expert-2
layer_range:
- 15
- 16
- sources:
- model: mixtral-expert-3
layer_range:
- 15
- 16
- sources:
- model: mixtral-expert-4
layer_range:
- 15
- 16
- sources:
- model: mixtral-expert-5
layer_range:
- 15
- 16
- sources:
- model: mixtral-expert-6
layer_range:
- 15
- 16
- sources:
- model: mixtral-expert-7
layer_range:
- 15
- 16
- sources:
- model: mixtral-expert-0
layer_range:
- 16
- 17
- sources:
- model: mixtral-expert-1
layer_range:
- 16
- 17
- sources:
- model: mixtral-expert-2
layer_range:
- 16
- 17
- sources:
- model: mixtral-expert-3
layer_range:
- 16
- 17
- sources:
- model: mixtral-expert-4
layer_range:
- 16
- 17
- sources:
- model: mixtral-expert-5
layer_range:
- 16
- 17
- sources:
- model: mixtral-expert-6
layer_range:
- 16
- 17
- sources:
- model: mixtral-expert-7
layer_range:
- 16
- 17
- sources:
- model: mixtral-expert-0
layer_range:
- 17
- 18
- sources:
- model: mixtral-expert-1
layer_range:
- 17
- 18
- sources:
- model: mixtral-expert-2
layer_range:
- 17
- 18
- sources:
- model: mixtral-expert-3
layer_range:
- 17
- 18
- sources:
- model: mixtral-expert-4
layer_range:
- 17
- 18
- sources:
- model: mixtral-expert-5
layer_range:
- 17
- 18
- sources:
- model: mixtral-expert-6
layer_range:
- 17
- 18
- sources:
- model: mixtral-expert-7
layer_range:
- 17
- 18
- sources:
- model: mixtral-expert-0
layer_range:
- 18
- 19
- sources:
- model: mixtral-expert-1
layer_range:
- 18
- 19
- sources:
- model: mixtral-expert-2
layer_range:
- 18
- 19
- sources:
- model: mixtral-expert-3
layer_range:
- 18
- 19
- sources:
- model: mixtral-expert-4
layer_range:
- 18
- 19
- sources:
- model: mixtral-expert-5
layer_range:
- 18
- 19
- sources:
- model: mixtral-expert-6
layer_range:
- 18
- 19
- sources:
- model: mixtral-expert-7
layer_range:
- 18
- 19
- sources:
- model: mixtral-expert-0
layer_range:
- 19
- 20
- sources:
- model: mixtral-expert-1
layer_range:
- 19
- 20
- sources:
- model: mixtral-expert-2
layer_range:
- 19
- 20
- sources:
- model: mixtral-expert-3
layer_range:
- 19
- 20
- sources:
- model: mixtral-expert-4
layer_range:
- 19
- 20
- sources:
- model: mixtral-expert-5
layer_range:
- 19
- 20
- sources:
- model: mixtral-expert-6
layer_range:
- 19
- 20
- sources:
- model: mixtral-expert-7
layer_range:
- 19
- 20
- sources:
- model: mixtral-expert-0
layer_range:
- 20
- 21
- sources:
- model: mixtral-expert-1
layer_range:
- 20
- 21
- sources:
- model: mixtral-expert-2
layer_range:
- 20
- 21
- sources:
- model: mixtral-expert-3
layer_range:
- 20
- 21
- sources:
- model: mixtral-expert-4
layer_range:
- 20
- 21
- sources:
- model: mixtral-expert-5
layer_range:
- 20
- 21
- sources:
- model: mixtral-expert-6
layer_range:
- 20
- 21
- sources:
- model: mixtral-expert-7
layer_range:
- 20
- 21
- sources:
- model: mixtral-expert-0
layer_range:
- 21
- 22
- sources:
- model: mixtral-expert-1
layer_range:
- 21
- 22
- sources:
- model: mixtral-expert-2
layer_range:
- 21
- 22
- sources:
- model: mixtral-expert-3
layer_range:
- 21
- 22
- sources:
- model: mixtral-expert-4
layer_range:
- 21
- 22
- sources:
- model: mixtral-expert-5
layer_range:
- 21
- 22
- sources:
- model: mixtral-expert-6
layer_range:
- 21
- 22
- sources:
- model: mixtral-expert-7
layer_range:
- 21
- 22
- sources:
- model: mixtral-expert-0
layer_range:
- 22
- 23
- sources:
- model: mixtral-expert-1
layer_range:
- 22
- 23
- sources:
- model: mixtral-expert-2
layer_range:
- 22
- 23
- sources:
- model: mixtral-expert-3
layer_range:
- 22
- 23
- sources:
- model: mixtral-expert-4
layer_range:
- 22
- 23
- sources:
- model: mixtral-expert-5
layer_range:
- 22
- 23
- sources:
- model: mixtral-expert-6
layer_range:
- 22
- 23
- sources:
- model: mixtral-expert-7
layer_range:
- 22
- 23
- sources:
- model: mixtral-expert-0
layer_range:
- 23
- 24
- sources:
- model: mixtral-expert-1
layer_range:
- 23
- 24
- sources:
- model: mixtral-expert-2
layer_range:
- 23
- 24
- sources:
- model: mixtral-expert-3
layer_range:
- 23
- 24
- sources:
- model: mixtral-expert-4
layer_range:
- 23
- 24
- sources:
- model: mixtral-expert-5
layer_range:
- 23
- 24
- sources:
- model: mixtral-expert-6
layer_range:
- 23
- 24
- sources:
- model: mixtral-expert-7
layer_range:
- 23
- 24
- sources:
- model: mixtral-expert-0
layer_range:
- 24
- 25
- sources:
- model: mixtral-expert-1
layer_range:
- 24
- 25
- sources:
- model: mixtral-expert-2
layer_range:
- 24
- 25
- sources:
- model: mixtral-expert-3
layer_range:
- 24
- 25
- sources:
- model: mixtral-expert-4
layer_range:
- 24
- 25
- sources:
- model: mixtral-expert-5
layer_range:
- 24
- 25
- sources:
- model: mixtral-expert-6
layer_range:
- 24
- 25
- sources:
- model: mixtral-expert-7
layer_range:
- 24
- 25
- sources:
- model: mixtral-expert-0
layer_range:
- 25
- 26
- sources:
- model: mixtral-expert-1
layer_range:
- 25
- 26
- sources:
- model: mixtral-expert-2
layer_range:
- 25
- 26
- sources:
- model: mixtral-expert-3
layer_range:
- 25
- 26
- sources:
- model: mixtral-expert-4
layer_range:
- 25
- 26
- sources:
- model: mixtral-expert-5
layer_range:
- 25
- 26
- sources:
- model: mixtral-expert-6
layer_range:
- 25
- 26
- sources:
- model: mixtral-expert-7
layer_range:
- 25
- 26
- sources:
- model: mixtral-expert-0
layer_range:
- 26
- 27
- sources:
- model: mixtral-expert-1
layer_range:
- 26
- 27
- sources:
- model: mixtral-expert-2
layer_range:
- 26
- 27
- sources:
- model: mixtral-expert-3
layer_range:
- 26
- 27
- sources:
- model: mixtral-expert-4
layer_range:
- 26
- 27
- sources:
- model: mixtral-expert-5
layer_range:
- 26
- 27
- sources:
- model: mixtral-expert-6
layer_range:
- 26
- 27
- sources:
- model: mixtral-expert-7
layer_range:
- 26
- 27
- sources:
- model: mixtral-expert-0
layer_range:
- 27
- 28
- sources:
- model: mixtral-expert-1
layer_range:
- 27
- 28
- sources:
- model: mixtral-expert-2
layer_range:
- 27
- 28
- sources:
- model: mixtral-expert-3
layer_range:
- 27
- 28
- sources:
- model: mixtral-expert-4
layer_range:
- 27
- 28
- sources:
- model: mixtral-expert-5
layer_range:
- 27
- 28
- sources:
- model: mixtral-expert-6
layer_range:
- 27
- 28
- sources:
- model: mixtral-expert-7
layer_range:
- 27
- 28
- sources:
- model: mixtral-expert-0
layer_range:
- 28
- 29
- sources:
- model: mixtral-expert-1
layer_range:
- 28
- 29
- sources:
- model: mixtral-expert-2
layer_range:
- 28
- 29
- sources:
- model: mixtral-expert-3
layer_range:
- 28
- 29
- sources:
- model: mixtral-expert-4
layer_range:
- 28
- 29
- sources:
- model: mixtral-expert-5
layer_range:
- 28
- 29
- sources:
- model: mixtral-expert-6
layer_range:
- 28
- 29
- sources:
- model: mixtral-expert-7
layer_range:
- 28
- 29
- sources:
- model: mixtral-expert-0
layer_range:
- 29
- 30
- sources:
- model: mixtral-expert-1
layer_range:
- 29
- 30
- sources:
- model: mixtral-expert-2
layer_range:
- 29
- 30
- sources:
- model: mixtral-expert-3
layer_range:
- 29
- 30
- sources:
- model: mixtral-expert-4
layer_range:
- 29
- 30
- sources:
- model: mixtral-expert-5
layer_range:
- 29
- 30
- sources:
- model: mixtral-expert-6
layer_range:
- 29
- 30
- sources:
- model: mixtral-expert-7
layer_range:
- 29
- 30
merge_method: passthrough
dtype: bfloat16
- Downloads last month
- 6
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.