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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 114223
    Prev
    Next
    INDEX
    Explanations

    themes related to filmmakers and their work

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

    films and foreign words

    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-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    conexion
    -0.35
     process
    -0.35
     Quarter
    -0.31
     patients
    -0.31
    Hochspringen
    -0.31
     Benefit
    -0.31
     while
    -0.30
     atraves
    -0.30
    Sec
    -0.30
    ↵
    -0.30
    POSITIVE LOGITS
     Winaray
    0.64
    +#+
    0.64
    CppMethod
    0.60
    SharedDtor
    0.60
     ConstraintSet
    0.60
    CodeAttribute
    0.57
    RTSN
    0.56
    ंदीखरीदारी
    0.56
     незавершена
    0.56
    ikusbot
    0.55
    Activations Density 0.038%

    No Known Activations