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
MLP Post
0-MLP-SM
27388
Prev
Next
MODEL
0-mlp-sm
Source/SAE
INDEX
Go
Explanations
instances of the token sequence starting with "[" and ending with "]"
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__mlp-sm_processed/0-mlp-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.0.hook_mlp_out
Hook Layer
0
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/0-mlp-sm/27388?embed=true&embedexplanation=true&embedplots=true&embedtest=true" title="Neuronpedia" style="height: 300px; width: 540px;"></iframe>
Link
https://www.neuronpedia.org/pythia-70m-deduped/0-mlp-sm/27388?embed=true&embedexplanation=true&embedplots=true&embedtest=true
Not in Any Lists
Add to List
▼
No Comments
ADD
Neuron Alignment
Index
Value
% of L₁
301
+0.14
0.8%
185
+0.11
0.6%
289
+0.10
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
354
+0.14
0.07
51
+0.11
0.05
188
+0.10
0.06
Negative Logits
İ
-2.06
Ļ
-1.91
Ĵ
-1.80
Ł
-1.77
¦
-1.76
į
-1.70
ª
-1.69
Ħ
-1.68
ĩ
-1.66
Ķ
-1.62
POSITIVE LOGITS
)\]
1.69
ourse
1.59
ergic
1.59
Mice
1.58
ous
1.54
lin
1.49
.-
1.48
ocial
1.47
®
1.45
omyces
1.44
Act
ivations
Density 0.039%
Stacked
Snippet
Full
Show Breaks
Hide Breaks
No Known Activations