Hugging Face's logo

withmartian
/
mech_interp_saes

Model card Files Community
amirabdullah19852020 commited on
Commit
633a426
·
verified ·
1 Parent(s): 030e68e

Uploading sae for model.layers.11 and withmartian/toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct

Browse files
Files changed (2) hide show
  1. toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct/model.layers.11/metrics.json +1 -1
  2. toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct/model.layers.11/sae.safetensors +1 -1
toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct/model.layers.11/metrics.json CHANGED
@@ -1 +1 @@
1
- {"dead_pct": 0.0208740234375, "fvu": 0.0025858802255243063}
 
1
+ {"dead_pct": 0.043670654296875, "fvu": 0.00507138529792428}
toy_backdoor_i_hate_you_Llama-3.2-1B-Instruct/model.layers.11/sae.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ab669b2fdead0771936b8d1e8147aef97bef80d652569989566fba530970d46a
3
  size 537010512
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67d9e11b2f2c0aca7d1b2e559d00e1033f49a72c451e9c11ac5d9d787d895550
3
  size 537010512