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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 13129
    Prev
    Next
    INDEX
    Explanations

    expressions of appreciation and attention to detail

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

    causal relationships expressed through words like "of," "to," and "with" that connect actions or outcomes to their preceding causes or agents.

    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
    ModelForm
    -0.39
    WebMethod
    -0.36
    ArgsConstructor
    -0.36
    ssohn
    -0.35
     acca
    -0.35
     proprie
    -0.35
     StatefulWidget
    -0.34
    endorff
    -0.34
    ModelSerializer
    -0.34
     Cadet
    -0.33
    POSITIVE LOGITS
     being
    0.76
     someone
    0.73
     having
    0.71
     him
    0.70
     people
    0.67
     us
    0.64
     somebody
    0.61
     alguém
    0.60
     adanya
    0.60
     seeing
    0.60
    Activations Density 0.603%

    No Known Activations