huypn16 commited on
Commit
547f968
1 Parent(s): 84cb311

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"sae": {"expansion_factor": 16, "normalize_decoder": true, "num_latents": 0, "k": 64, "multi_topk": false}, "batch_size": 32, "grad_acc_steps": 1, "micro_acc_steps": 1, "lr": null, "lr_warmup_steps": 1000, "auxk_alpha": 0.0, "dead_feature_threshold": 10000000, "hookpoints": ["layers.8", "layers.10", "layers.12", "layers.14", "layers.16", "layers.18"], "layers": [8, 10, 12, 14, 16, 18], "layer_stride": 1, "distribute_modules": true, "save_every": 500, "log_to_wandb": true, "run_name": "qwen-math", "wandb_log_frequency": 1, "model": "Qwen/Qwen2.5-1.5B-Instruct", "dataset": "huypn16/MetaMath-SAE", "split": "train", "ctx_len": 1024, "hf_token": null, "load_in_8bit": false, "max_examples": null, "resume": false, "seed": 42, "data_preprocessing_num_proc": 112}
layers.10/cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"expansion_factor": 16, "normalize_decoder": true, "num_latents": 0, "k": 64, "multi_topk": false, "d_in": 1536}
layers.10/sae.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:960287254a58b3d8161b2611986ab1e42270d85c90ff11ff545a55612ec3a213
3
+ size 302094672
layers.12/cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"expansion_factor": 16, "normalize_decoder": true, "num_latents": 0, "k": 64, "multi_topk": false, "d_in": 1536}
layers.12/sae.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b673f43fbb9ffb919e5f3ff7694ca0288a6329bfaa44d493ed06deaf7c146931
3
+ size 302094672
layers.14/cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"expansion_factor": 16, "normalize_decoder": true, "num_latents": 0, "k": 64, "multi_topk": false, "d_in": 1536}
layers.14/sae.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:133296d4c1b17a951833166bb31f67cbcf92972025e4156a592095728dcd519b
3
+ size 302094672
layers.16/cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"expansion_factor": 16, "normalize_decoder": true, "num_latents": 0, "k": 64, "multi_topk": false, "d_in": 1536}
layers.16/sae.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbc7bbc68ffdc8ef50bc849535c5b02ee6935af383d24f2f01c880d0d5952be3
3
+ size 302094672
layers.18/cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"expansion_factor": 16, "normalize_decoder": true, "num_latents": 0, "k": 64, "multi_topk": false, "d_in": 1536}
layers.18/sae.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:167a82ea402b3d4032c043402eae58b9bed2ac0313a3e641295044ab8566e62d
3
+ size 302094672
layers.8/cfg.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"expansion_factor": 16, "normalize_decoder": true, "num_latents": 0, "k": 64, "multi_topk": false, "d_in": 1536}
layers.8/sae.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18dc8eafcf7d71b38d48a8d3207071e2199c7c2eb590c5eeb57c9f221dc69c24
3
+ size 302094672
lr_scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4414c798ad39f32ec4002492a37d643b1410efa070c8be1d15ef81cb27b1ab7d
3
+ size 1076
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94a452ffde9c29650301bab0ed8b695646101a9d92747818b74f97136b8eab01
3
+ size 302713856
state.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8665b917841a3c7a4c6b4d14436a6427ebff20ecdf4ef910d10f18882c2e4d25
3
+ size 394572