RealmSky/layerskip-llama3.2-1B
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: RealmSky/layerskip-llama3.2-1B
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-09-26 10:05:13 +08:00
commit 4bd23626bd
18 changed files with 131017 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
original/consolidate_params.json Normal file
View File

@@ -0,0 +1,7 @@
{
"dtype": "bf16",
"model_parallel_size": 1,
"on_gpu": true,
"src": "/fsx-atom/melhoushi/xldumps/continual_llama3.2_1B_gpus_64_shared_emb_higher_lr_ee_0.1_ld_0.1_128_gpus/continual_llama3.2_1B_gpus_64_shared_emb_higher_lr_ee_0.1_ld_0.1_128_gpus_run000/checkpoints/checkpoint_0100000/",
"tgt": "/fsx-atom/melhoushi/xldumps/continual_llama3.2_1B_gpus_64_shared_emb_higher_lr_ee_0.1_ld_0.1_128_gpus/continual_llama3.2_1B_gpus_64_shared_emb_higher_lr_ee_0.1_ld_0.1_128_gpus_run000/checkpoints/checkpoint_0100000_consolidated/"
}