Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 37-GEMMASCOPE-RES-16K
    6. 3612
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter "th" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.37.hook_resid_post
    Hook Layer
    37
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AndEndTag
    -0.62
    rungsseite
    -0.60
    jewództ
    -0.47
    setVerticalGroup
    -0.44
     ángeles
    -0.40
     steder
    -0.40
     mówią
    -0.39
    madeus
    -0.38
     mało
    -0.38
     émer
    -0.38
    POSITIVE LOGITS
     مرئيه
    0.44
    createServer
    0.43
    oredCriteria
    0.43
    reonine
    0.42
    reed
    0.41
    elemField
    0.40
    rought
    0.39
     intptr
    0.39
    webElementGuid
    0.39
    aliana
    0.38
    Activations Density 0.215%

    No Known Activations