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. 33-GEMMASCOPE-RES-131K
    6. 65940
    Prev
    Next
    INDEX
    Explanations

    references to cellular components and their localization

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_33/width_131k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.33.hook_resid_post
    Hook Layer
    33
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    setVerticalGroup
    -0.70
     Infórmanos
    -0.69
     informée
    -0.63
     Roskov
    -0.61
    __(/*!
    -0.61
     GenerationType
    -0.60
     للمعارف
    -0.60
     nakalista
    -0.59
    Tikang
    -0.58
     AssemblyCulture
    -0.57
    POSITIVE LOGITS
     Space
    0.56
     Spaces
    0.53
     SPACE
    0.50
     space
    0.49
    Space
    0.49
     空
    0.45
     spaces
    0.44
    空
    0.43
    Spaces
    0.43
     vacancy
    0.42
    Activations Density 0.004%

    No Known Activations