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-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 955
    Prev
    Next
    INDEX
    Explanations

    code structures and their associated elements

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

    C code structures and bitfields

    np_acts-logits-general · gemini-2.0-flash

    various languages and code structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
    ----</
    -0.80
     betweenstory
    -0.79
     snippetHide
    -0.79
     للمعارف
    -0.76
    parsedMessage
    -0.75
    WireFormatLite
    -0.68
    richTextPanel
    -0.67
    URLException
    -0.67
    StoryboardSegue
    -0.67
    providedIn
    -0.66
    POSITIVE LOGITS
     feroit
    0.32
     auroit
    0.31
     pourriez
    0.31
     pouvoit
    0.31
     trozos
    0.30
     Grecs
    0.27
     ciudadana
    0.27
     mètre
    0.27
    こちらも
    0.26
     tremp
    0.26
    Activations Density 0.217%

    No Known Activations