yuhengtu-bytedance/DataDecide-fineweb-pro-1B-60000_62500_65000_67500_69369_weightedavg_merge
0252
DataDecide-fineweb-pro-1B-6000062500650006750069369weightedavgmerge
This is a merge of pre-trained language models created using mergekit.
Merge Details
Merge Method
This model was merged using the Linear merge method using /opt/tiger/PanSafetyBetterMeasurement/mergescalingckptscache/source_ckpts/fineweb-pro/step69369 as a base.
Models Merged
The following models were included in the merge:
- /opt/tiger/PanSafetyBetterMeasurement/mergescalingckptscache/source_ckpts/fineweb-pro/step60000
- /opt/tiger/PanSafetyBetterMeasurement/mergescalingckptscache/source_ckpts/fineweb-pro/step65000
- /opt/tiger/PanSafetyBetterMeasurement/mergescalingckptscache/source_ckpts/fineweb-pro/step62500
- /opt/tiger/PanSafetyBetterMeasurement/mergescalingckptscache/source_ckpts/fineweb-pro/step67500
Configuration
The following YAML configuration was used to produce this model:
models:
- model: /opt/tiger/Pan_Safety_Better_Measurement/merge_scaling_ckpts_cache/source_ckpts/fineweb-pro/step60000
parameters:
weight: 1
- model: /opt/tiger/Pan_Safety_Better_Measurement/merge_scaling_ckpts_cache/source_ckpts/fineweb-pro/step62500
parameters:
weight: 2
- model: /opt/tiger/Pan_Safety_Better_Measurement/merge_scaling_ckpts_cache/source_ckpts/fineweb-pro/step65000
parameters:
weight: 3
- model: /opt/tiger/Pan_Safety_Better_Measurement/merge_scaling_ckpts_cache/source_ckpts/fineweb-pro/step67500
parameters:
weight: 4
- model: /opt/tiger/Pan_Safety_Better_Measurement/merge_scaling_ckpts_cache/source_ckpts/fineweb-pro/step69369
parameters:
weight: 5
base_model: /opt/tiger/Pan_Safety_Better_Measurement/merge_scaling_ckpts_cache/source_ckpts/fineweb-pro/step69369
merge_method: linear
parameters:
normalize: true
dtype: float32
out_dtype: bfloat16
