Neuronpedia
Get Started
API
Releases
Jump To
Search
Models
Circuit Tracer
NEW
Steer
SAE Evals
Exports
Slack
Blog
Privacy & Terms
Contact
Sign In
© Neuronpedia 2025
Privacy & Terms
Blog
GitHub
Slack
Twitter
Contact
Home
Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
Llama3.1-8B-IT (Instruct)
Resid Post - 131k
11-RESID-POST-AA
130659
Prev
Next
MODEL
11-resid-post-aa
Source/SAE
INDEX
Go
Explanations
us
np_max-act · gemini-2.0-flash
No Scores
The neuron detects occurrences of the word “us,” especially in calls to action (e.g. “contact us,” “help us,” “mail us,” “visit us”).
oai_token-act-pair · o4-mini
Triggered by @xinyanhu8
No Scores
New Auto-Interp
AutoInterp Type
claude-4-5-haiku
Generate
Top Features by Cosine Similarity
Configuration
andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
How To Load
Dataset (Dashboard)
Various
Features
131,072
Data Type
float32
Hook Name
blocks.11.hook_resid_post
Architecture
standard
Context Size
1,024
Dataset
monology/pile-uncopyrighted
Show All
Embeds
Plots
Explanation
Show Test Field
Default Test Text
IFrame
<iframe src=https://www.neuronpedia.org/llama3.1-8b-it/11-resid-post-aa/130659?embed=true&embedexplanation=true&embedplots=true&embedtest=true" title="Neuronpedia" style="height: 300px; width: 540px;"></iframe>
Link
https://www.neuronpedia.org/llama3.1-8b-it/11-resid-post-aa/130659?embed=true&embedexplanation=true&embedplots=true&embedtest=true
Not in Any Lists
Add to List
▼
No Comments
ADD
Negative Logits
-di
-0.08
terní
-0.08
Ele
-0.07
Day
-0.07
гр
-0.07
její
-0.07
(feed
-0.07
přím
-0.07
rhe
-0.07
champagne
-0.07
POSITIVE LOGITS
us
0.18
Us
0.14
us
0.12
us
0.11
US
0.11
Us
0.10
US
0.10
AS
0.10
(us
0.09
-us
0.09
Act
ivations
Density 0.031%
Test
Steer
Stacked
Snippet
Full
Show Raw Tokens
Show Formatted
Show Breaks
Hide Breaks
No Known Activations
