© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 16262
    Prev
    Next
    INDEX
    Explanations

    expressions and emotional nuances of characters in dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    parsedMessage
    -0.88
    aarrggbb
    -0.85
    Portale
    -0.85
    <pad>
    -0.83
    <unused43>
    -0.82
    <unused41>
    -0.82
    <unused74>
    -0.82
     AssemblyCulture
    -0.82
    <unused8>
    -0.82
    <unused23>
    -0.82
    POSITIVE LOGITS
     face
    0.56
     smile
    0.50
     faces
    0.48
     frown
    0.46
     frowning
    0.46
     look
    0.44
     frowns
    0.42
    face
    0.41
     smiling
    0.40
     sonrisa
    0.39
    Activations Density 0.065%

    No Known Activations