Downloads · 30 days
9
32% of all-time downloads
wudi_gemma_fp32_1_iter_25
tuanpasg/wudi_gemma_fp32_1_iter_25
wudi_gemma_fp32_1_iter_25 is a machine learning model from tuanpasg. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
- Base model: google/gemma-2-2b - Algorithm: wudimerge - Save path: /media02/nthuy08/outs/wudigemmafp321iter25 - Fine-tuned checkpoints: ['MergeBench/gemma-2-2binstruction', 'MergeBench/gemma-2-2bmath', 'MergeBench/ge…
All-time downloads
28
Public
Parameters
2.6B
5.3 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors5.2 GB · 99%
At a glance
- Model type
- gemma2
- Access
- Public
- Created
- Sep 6, 2026
- Updated
- Sep 6, 2026
- SHA
- 8a8034cf
Model card
From the Hugging Face model README
Merged Model
- Base model: google/gemma-2-2b
- Algorithm:
wudi_merge - Save path:
/media02/nthuy08/outs/wudi_gemma_fp32_1_iter_25 - Fine-tuned checkpoints: ['MergeBench/gemma-2-2b_instruction', 'MergeBench/gemma-2-2b_math', 'MergeBench/gemma-2-2b_coding']
- Merge runtime (s): 344.303
Arguments
{
"out": "/media02/nthuy08/outs/wudi_gemma_fp32_1_iter_25",
"scaling": 1.0,
"dtype": "bfloat16",
"device_map": "cpu",
"merge_method": "wudi_merge",
"exclude": [
".*embed.*",
".*lm_head.*"
],
"wudi_variant": "wudi_all_linear",
"wudi_iter": 25,
"wudi_lr": 1e-05,
"wudi_weight_decay": 0.0,
"wudi_alpha": 0.0,
"wudi_device": "cuda",
"wudi_warm_start": false,
"wudi_cfs_ridge": 0.0,
"wudi_fallback": "task_arithmetic",
"wudi_fallback_scaling": 1.0,
"wudi_K": 0.7,
"wudi_sparsify_variant": "ties_sparsify",
"wudi_loss_log_csv": null,
"effective_exclude": [
".*embed_tokens\\.weight$",
".*lm_head\\.weight$",
".*embed.*",
".*lm_head.*"
]
}
- Type
- gemma2
- Created
- Sep 6, 2026
- Updated
- Sep 6, 2026