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
    4. Residual Stream - 131k
    5. 32-GEMMASCOPE-RES-131K
    6. 130038
    Prev
    Next
    INDEX
    Explanations

    adjectives that convey positive qualities or capabilities

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

    adjective as

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.32.hook_resid_post
    Hook Layer
    32
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    httphttps
    -0.73
    contentLoaded
    -0.61
     kaarangay
    -0.56
     ErrIntOverflow
    -0.55
     ſei
    -0.52
    ]")]
    -0.50
     beſte
    -0.50
     faſt
    -0.50
     المعيارى
    -0.50
    újo
    -0.49
    POSITIVE LOGITS
     as
    2.16
     sebagai
    1.09
     как
    1.09
     kuin
    1.09
     як
    1.03
     quanto
    0.95
     như
    0.92
     както
    0.85
     كما
    0.82
     Sebagai
    0.79
    Activations Density 0.109%

    No Known Activations