Mistral-Nemo-Base-2407-20b / mergekit_config.yml
win10's picture
Upload folder using huggingface_hub
a866406 verified
dtype: bfloat16
merge_method: passthrough
slices:
- sources:
- layer_range: [0, 2]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [1, 3]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [2, 4]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [3, 5]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
# 以下是新增的層
- sources:
- layer_range: [4, 6]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [5, 7]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [6, 8]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [7, 9]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [8, 10]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [9, 11]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [10, 12]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [11, 13]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [12, 14]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [13, 15]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [14, 16]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [15, 17]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [16, 18]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [17, 19]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [18, 20]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [19, 21]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [20, 22]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [21, 23]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [22, 24]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [23, 25]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [24, 26]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [25, 27]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [26, 28]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [27, 29]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [28, 30]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [29, 31]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [30, 32]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [31, 33]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [32, 34]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [33, 35]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [34, 36]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [35, 37]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [36, 38]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [37, 39]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [38, 40]
model: unsloth/Mistral-Nemo-Base-2407
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0