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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 65635
    Prev
    Next
    INDEX
    Explanations

    phrases indicating enthusiasm or strong interest

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

    intensity of feeling or action

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    LookAnd
    -0.60
    exitRule
    -0.59
    providedIn
    -0.59
     Situations
    -0.59
     Cabello
    -0.57
    prnewswire
    -0.57
    COA
    -0.57
     Clothes
    -0.56
     تكبرها
    -0.56
    bootstrapcdn
    -0.56
    POSITIVE LOGITS
     avid
    0.88
     Avid
    0.66
     passionate
    0.56
     enthusiastic
    0.55
     active
    0.54
    active
    0.49
     apasion
    0.47
    Active
    0.47
     activo
    0.47
     actively
    0.46
    Activations Density 0.002%

    No Known Activations