© 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-2-2B
    4. Residual Stream - 16k
    5. 22-GEMMASCOPE-RES-16K
    6. 13577
    Prev
    Next
    INDEX
    Explanations

    code with function declarations and data manipulation operations

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

    Code snippets and image descriptions

    np_acts-logits-general · gemini-2.0-flash

    image focusing, periodic review, ; CNP

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    addGap
    -0.62
    uffff
    -0.58
    DebuggerNonUser
    -0.56
    onite
    -0.51
     cards
    -0.51
    cards
    -0.51
     مشين
    -0.50
    verifyException
    -0.49
    ')));
    -0.48
     Cards
    -0.48
    POSITIVE LOGITS
    dispatch
    0.56
    [[
    0.55
     UIImage
    0.55
    Notae
    0.53
     [[
    0.52
    ︎
    0.51
     nil
    0.50
    aarrggbb
    0.49
    anyol
    0.49
    NSLog
    0.49
    Activations Density 0.043%

    No Known Activations