© 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-IT
    4. Residual Stream - 131k
    5. 31-GEMMASCOPE-RES-131K
    6. 46869
    Prev
    Next
    INDEX
    Explanations

    references to heads and their positioning or actions

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

    physical and metaphorical head

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    AndEndTag
    -0.58
    awtextra
    -0.56
     Wikiseite
    -0.49
     âmes
    -0.48
    rativa
    -0.47
     informée
    -0.46
     TextAppearance
    -0.46
    enderror
    -0.46
    Personendaten
    -0.45
    reactstrap
    -0.44
    POSITIVE LOGITS
     head
    0.47
     Head
    0.44
     HEAD
    0.40
    Head
    0.39
    head
    0.37
    ########.
    0.36
    SuspendLayout
    0.35
    toThrow
    0.35
     swirling
    0.34
    toMatchSnapshot
    0.34
    Activations Density 0.009%

    No Known Activations