Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 25-GEMMASCOPE-RES-16K
    6. 8216
    Prev
    Next
    INDEX
    Explanations

    quotes or phrases that indicate direct speech or dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    adaptiveStyles
    -0.63
    awtextra
    -0.61
    setOnAction
    -0.60
    extAlignment
    -0.58
    marginHorizontal
    -0.57
    orithmic
    -0.56
    luca
    -0.55
     kaynağından
    -0.54
    numRows
    -0.53
    asString
    -0.53
    POSITIVE LOGITS
    get
    0.93
    have
    0.88
    make
    0.84
    take
    0.83
    raise
    0.82
    try
    0.82
    do
    0.80
    give
    0.80
    go
    0.78
    hold
    0.78
    Activations Density 0.385%

    No Known Activations