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
Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
GPT2-Small
Residual Stream
0-RES-JB
8944
Prev
Next
MODEL
0-res-jb
Source/SAE
INDEX
Go
Explanations
proper nouns and numerical values
oai_token-act-pair · gpt-3.5-turbo
No Scores
a repeated or emphasized special character or symbol
oai_token-act-pair · gpt-4o-mini
Triggered by @bot
No Scores
New Auto-Interp
AutoInterp Type
claude-3-5-haiku-20241022
Generate
Top Features by Cosine Similarity
Comparing With
GPT2-SMALL @ 0-res-jb
Configuration
jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
How To Load
Prompts (Dashboard)
24,576 prompts, 128 tokens each
Dataset (Dashboard)
Skylion007/openwebtext
Features
24,576
Data Type
torch.float32
Hook Point
blocks.0.hook_resid_pre
Architecture
standard
Context Size
128
Dataset
Skylion007/openwebtext
Hook Point Layer
0
Activation Function
relu
Show All
Embeds
Plots
Explanation
Show Test Field
Default Test Text
IFrame
<iframe src=https://www.neuronpedia.org/gpt2-small/0-res-jb/8944?embed=true&embedexplanation=true&embedplots=true&embedtest=true" title="Neuronpedia" style="height: 300px; width: 540px;"></iframe>
Link
https://www.neuronpedia.org/gpt2-small/0-res-jb/8944?embed=true&embedexplanation=true&embedplots=true&embedtest=true
Not in Any Lists
Add to List
▼
No Comments
ADD
Negative Logits
demand
-0.69
emphasis
-0.68
market
-0.67
Equ
-0.64
ised
-0.64
syndrome
-0.63
ises
-0.61
mix
-0.59
suit
-0.59
board
-0.59
POSITIVE LOGITS
Ļ
4.43
Ľ
1.89
ħ
1.88
ĺ
1.87
©
1.75
ij
1.72
ļ
1.71
Ĺ
1.71
¤
1.67
ķ
1.67
Act
ivations
Density 0.585%
Test
Steer
Stacked
Snippet
Full
Show Breaks
Hide Breaks
No Known Activations
