© Neuronpedia 2026
Privacy & Terms
Blog
GitHub
Slack
Twitter
Contact
Neuronpedia
Jacobian Lens
NEW
Natural Language
Autoencoders
NEW
Assistant Axis
NEW
Circuit Tracer
UPDATE
Releases
Jump To
Search
Steer
SAE Evals
Exports
interp-engine
NEW
Guides
API
Community
Blog
Privacy & Terms
Contact
Sign In
Home
Gemma-2-27B
34-GEMMASCOPE-RES-131K
117511
Prev
Next
MODEL
34-gemmascope-res-131k
Source/SAE
INDEX
Go
Explanations
school types, disease types, weapon types, experiment types
np_acts-logits-general · gemini-2.5-flash-lite
No Scores
descriptions of institutional or scientific classifications with multiple attributes.
oai_token-act-pair · claude-3-7-sonnet-20250219
Triggered by @neilrathi
No Scores
The neuron fires on descriptive modifiers—especially compound or category‐defining adjectives and nouns (e.g. “private,” “educational,” “Catholic,” “co-educational,” “two-component,” “hard court”) that qualify or classify the main noun.
oai_token-act-pair · o4-mini
Triggered by @jyhe0408
No Scores
New Auto-Interp
AutoInterp Type
claude-4-5-haiku
Generate
Top Features by Cosine Similarity
Configuration
google/gemma-scope-27b-pt-res/layer_34/width_131k
Prompts (Dashboard)
24,576 prompts, 128 tokens each
Dataset (Dashboard)
monology/pile-uncopyrighted
No Configuration Found
Show All
Embeds
Show Plots
Show Explanation
Show Activations
Show Test Field
Show Steer
Show Link
IFrame
<iframe src="https://www.neuronpedia.org/gemma-2-27b/34-gemmascope-res-131k/117511?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/gemma-2-27b/34-gemmascope-res-131k/117511?embed=true&embedexplanation=true&embedplots=true&embedsteer=true&embedactivations=true&embedlink=true&embedtest=true
Not in Any Lists
Add to List
▼
No Comments
ADD
Negative Logits
為に
-1.52
マジで
-1.38
ようになって
-1.35
凄く
-1.33
感じで
-1.31
ことは
-1.30
んですよ
-1.30
新しく
-1.29
ときは
-1.28
場所で
-1.27
POSITIVE LOGITS
<bos>
1.63
しかも
1.38
その
1.37
を
1.29
もちゃ
1.29
дован
1.23
などが
1.21
мои
1.16
もしろ
1.14
且
1.14
Act
ivations
Density 0.687%
Stacked
Snippet
Full
Show Breaks
Hide Breaks
No Known Activations