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. 22-GEMMASCOPE-RES-16K
    6. 11166
    Prev
    Next
    INDEX
    Explanations

    instances of the word "action" and its variants in various contexts related to activities or operations

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

    "Action" or "action"

    np_acts-logits-general · gemini-2.0-flash

    action

    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
     nahilalakip
    -0.93
     GenerationType
    -0.84
     дописавши
    -0.70
    ]--;
    -0.68
    WriteTagHelper
    -0.68
    ConstraintMaker
    -0.66
     ostavi
    -0.63
    InSection
    -0.62
    parsedMessage
    -0.62
     estekak
    -0.61
    POSITIVE LOGITS
    getAction
    0.73
     afirm
    0.72
     UIAlert
    0.72
     ACTION
    0.71
     Actions
    0.71
    //
    0.69
    actions
    0.69
     Action
    0.68
    ACTION
    0.68
    Actions
    0.68
    Activations Density 0.102%

    No Known Activations