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. 129897
    Prev
    Next
    INDEX
    Explanations

    structures and materials related to craftsmanship and construction

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

    made of material

    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
     createSprite
    -0.65
    tagHelperRunner
    -0.65
    tagext
    -0.60
     puentes
    -0.59
    AndEndTag
    -0.56
    RenderAtEndOf
    -0.56
     bridges
    -0.54
    Pile
    -0.52
     doccia
    -0.52
     незавершена
    -0.52
    POSITIVE LOGITS
     pure
    0.48
     stainless
    0.41
     untreated
    0.40
     native
    0.38
     recycled
    0.38
     purified
    0.38
    pure
    0.37
     natural
    0.36
     Douglas
    0.36
     white
    0.36
    Activations Density 0.036%

    No Known Activations