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. 2632
    Prev
    Next
    INDEX
    Explanations

    references to academic publications and articles

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

    academic and legal citations

    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
     Christi
    -0.32
     gynhyrchwyd
    -0.32
     reuni
    -0.32
    Bert
    -0.31
     Reunion
    -0.31
    ometer
    -0.30
    BufferedReader
    -0.30
     Bert
    -0.30
    берт
    -0.30
    Intern
    -0.30
    POSITIVE LOGITS
    complexContent
    0.66
    principalColumn
    0.65
    RUnlock
    0.63
     يتيمه
    0.58
    RTHOOK
    0.58
    endphp
    0.57
    ########.
    0.57
    MessageState
    0.57
     disponibilités
    0.57
     Biôgrafia
    0.56
    Activations Density 0.042%

    No Known Activations