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 - 16k
    5. 9-GEMMASCOPE-RES-16K
    6. 3769
    Prev
    Next
    INDEX
    Explanations

    closing HTML tags

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

    HTML list items (`<li>`)

    np_acts-logits-general · gemini-2.0-flash

    list and link tags

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     nélkül
    -0.42
     Exteriores
    -0.39
     Blatt
    -0.36
     build
    -0.36
     civilización
    -0.35
    sourceType
    -0.34
     Grüßen
    -0.33
     examine
    -0.33
     isso
    -0.33
    SCRIBE
    -0.33
    POSITIVE LOGITS
     kasarigan
    0.70
    </
    0.68
    transQ
    0.65
    -------------</
    0.65
    MigrationBuilder
    0.63
     surla
    0.61
    __":
    
    0.60
    tvguidetime
    0.60
    fromnode
    0.60
    ]")]
    0.59
    Activations Density 0.118%

    No Known Activations