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. 80569
    Prev
    Next
    INDEX
    Explanations

    elements related to functions, classifications, and results in academic or technical discussions

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

    based on, being sarcastic, justice, successful building, role, model, shed light

    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
    findpost
    -0.50
     invokingState
    -0.50
    ymce
    -0.46
    Jeografia
    -0.45
    ftagPool
    -0.45
    writeFieldEnd
    -0.45
     صوتيه
    -0.43
     فريبيس
    -0.43
    )_/¯
    -0.43
    WithIOException
    -0.43
    POSITIVE LOGITS
    новништво
    0.42
    IsUnicode
    0.40
    __':
    
    0.39
    InteropServices
    0.38
     comigo
    0.38
     Works
    0.37
    centerTitle
    0.37
    onomian
    0.37
    DECREF
    0.37
    Folgende
    0.36
    Activations Density 0.194%

    No Known Activations