© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 2715
    Prev
    Next
    INDEX
    Explanations

    syntactical elements and control flow keywords in programming code

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    loop increment operators and counter variable updates in code.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     „
    -0.50
    „
    -0.44
    RTLI
    -0.39
     („
    -0.36
    permitAll
    -0.35
    ‟
    -0.34
    |_
    -0.33
    ^)
    -0.32
    dragen
    -0.32
    _{[
    -0.32
    POSITIVE LOGITS
    ++
    1.57
     ++
    1.52
     (++
    1.34
    ++$
    1.23
    ++,
    1.23
     increment
    1.21
    ++;
    1.19
     increase
    1.18
     increasing
    1.18
     INCRE
    1.16
    Activations Density 0.349%

    No Known Activations