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
3-RES-JB
10171
Prev
Next
MODEL
3-res-jb
Source/SAE
INDEX
Go
Explanations
numerical values, particularly occurrences of the number 2009
oai_token-act-pair · gpt-3.5-turbo
No Scores
references to the number 200
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 @ 3-res-jb
Configuration
jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
Architecture
standard
Context Size
128
Dataset
Skylion007/openwebtext
Hook Point Layer
3
Activation Function
relu
Show All
Embeds
Plots
Explanation
Show Test Field
Default Test Text
IFrame
<iframe src=https://www.neuronpedia.org/gpt2-small/3-res-jb/10171?embed=true&embedexplanation=true&embedplots=true&embedtest=true" title="Neuronpedia" style="height: 300px; width: 540px;"></iframe>
Link
https://www.neuronpedia.org/gpt2-small/3-res-jb/10171?embed=true&embedexplanation=true&embedplots=true&embedtest=true
Not in Any Lists
Add to List
▼
No Comments
ADD
Negative Logits
warts
-0.80
book
-0.79
aken
-0.73
kell
-0.73
robe
-0.71
stal
-0.70
Äĩ
-0.69
cone
-0.67
achi
-0.66
isen
-0.64
POSITIVE LOGITS
00000
0.95
Hz
0.94
mph
0.90
ILCS
0.80
kHz
0.78
kHz
0.77
MHz
0.76
mph
0.74
Leban
0.74
Hz
0.73
Act
ivations
Density 0.016%
Test
Steer
Stacked
Snippet
Full
Show Breaks
Hide Breaks
No Known Activations
