© Neuronpedia 2026
Privacy & Terms
Blog
GitHub
Slack
Twitter
Contact
Neuronpedia
Get Started
API
Releases
Jump To
Search
Models
Assistant Axis
NEW
Circuit Tracer
NEW
Steer
SAE Evals
Exports
Community
Blog
Privacy & Terms
Contact
Sign In
PYTHIA-70M-DEDUPED · 2-MLP-SM · 26619 | Neuronpedia
Home
Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
Pythia-70M-Deduped
MLP Post
2-MLP-SM
26619
Prev
Next
MODEL
2-mlp-sm
Source/SAE
INDEX
Go
Explanations
code structure and closure statements
oai_token-act-pair · gpt-4o-mini
Triggered by @bot
No Scores
New Auto-Interp
AutoInterp Type
claude-4-5-haiku
Generate
Top Features by Cosine Similarity
Configuration
ctigges/pythia-70m-deduped__mlp-sm_processed/2-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.2.hook_mlp_out
Hook Layer
2
Architecture
standard
Context Size
128
Dataset
EleutherAI/the_pile_deduplicated
Activation Function
relu
Show All
Embeds
Show Plots
Show Explanation
Show Activations
Show Test Field
Show Steer
Show Link
IFrame
<iframe src="https://www.neuronpedia.org/pythia-70m-deduped/2-mlp-sm/26619?embed=true&embedexplanation=true&embedplots=true&embedsteer=true&embedactivations=true&embedlink=true&embedtest=true" title="Neuronpedia" style="height: 300px; width: 540px;"></iframe>
Link
https://www.neuronpedia.org/pythia-70m-deduped/2-mlp-sm/26619?embed=true&embedexplanation=true&embedplots=true&embedsteer=true&embedactivations=true&embedlink=true&embedtest=true
Not in Any Lists
Add to List
▼
No Comments
ADD
Neuron Alignment
Index
Value
% of L₁
424
+0.14
0.8%
326
+0.12
0.6%
265
+0.12
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
414
+0.14
0.04
159
+0.12
0.05
188
+0.12
0.04
Negative Logits
>();
-1.70
ently
-1.60
=".
-1.56
YRIGHT
-1.52
>';
-1.48
=$
-1.47
amines
-1.40
)){
-1.39
'</
-1.39
=#
-1.39
POSITIVE LOGITS
¹
2.32
ĨĴ
2.21
IJ
2.15
ĸ´
2.08
Ĥ¬
2.08
Ŀ
2.05
¶
2.04
¬
2.02
·
2.01
¸
2.01
Act
ivations
Density 0.220%
Stacked
Snippet
Full
Show Raw Tokens
Show Formatted
Show Breaks
Hide Breaks
No Known Activations