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
Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
Pythia-70M-Deduped
Residuals
2-RES-SM
29358
Prev
Next
MODEL
2-res-sm
Source/SAE
INDEX
Go
Explanations
phrases indicating specific conditions or characteristics
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
Configuration
ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
How To Load
Prompts (Dashboard)
32,768 prompts, 128 tokens each
Dataset (Dashboard)
monology/pile-uncopyrighted
Features
32,768
Data Type
torch.float32
Hook Name
blocks.2.hook_resid_post
Hook Layer
2
Architecture
standard
Context Size
128
Dataset
EleutherAI/the_pile_deduplicated
Activation Function
relu
Show All
Embeds
Plots
Explanation
Show Test Field
Default Test Text
IFrame
<iframe src=https://www.neuronpedia.org/pythia-70m-deduped/2-res-sm/29358?embed=true&embedexplanation=true&embedplots=true&embedtest=true" title="Neuronpedia" style="height: 300px; width: 540px;"></iframe>
Link
https://www.neuronpedia.org/pythia-70m-deduped/2-res-sm/29358?embed=true&embedexplanation=true&embedplots=true&embedtest=true
Not in Any Lists
Add to List
▼
No Comments
ADD
Neuron Alignment
Index
Value
% of L₁
156
+0.18
1.0%
125
+0.12
0.7%
369
+0.11
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
454
+0.18
0.04
125
+0.12
0.04
244
+0.11
0.06
Negative Logits
table
-1.57
feas
-1.52
tables
-1.49
ible
-1.43
...\...\
-1.40
ersion
-1.39
irector
-1.37
icio
-1.37
не
-1.37
orld
-1.34
POSITIVE LOGITS
Ī
2.96
↵
2.88
↵
2.88
2.88
↵
2.88
↵
2.88
č↵
2.88
2.88
↵↵
2.88
č↵
2.88
Act
ivations
Density 0.494%
Stacked
Snippet
Full
Show Breaks
Hide Breaks
No Known Activations
