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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 1777
    Prev
    Next
    INDEX
    Explanations

    phrases related to recipes and cooking methods

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

    numerical phrases and lists (like "5 Ways", "21 Reasons", "9 times").

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    list-style, count-oriented headlines and phrases that signal enumerations or quantified claims (e.g., numbered “reasons/ways/times” and similar listicle structures).

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    list items or enumerated points, particularly in the context of titles or headings that begin with a number.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    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
    
    -0.58
    principalColumn
    -0.57
     dõi
    -0.54
    esModule
    -0.52
    MigrationBuilder
    -0.51
    /*
    -0.45
     ""],
    -0.44
    EndContext
    -0.44
    StoryboardSegue
    -0.43
    ◀
    -0.42
    POSITIVE LOGITS
    Top
    0.41
    يكب
    0.40
    ofe
    0.40
    قایناق‌لار
    0.40
     Reasons
    0.39
     hydrothermal
    0.39
    Tech
    0.39
     sacked
    0.38
     Ways
    0.38
     <=",
    0.38
    Activations Density 0.083%

    No Known Activations