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. 9-GEMMASCOPE-RES-16K
    6. 11268
    Prev
    Next
    INDEX
    Explanations

    instances of contrasting or comparative relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 9-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_9/width_16k/average_l0_100
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    AutoModerator
    -0.49
     defaultstate
    -0.43
    ſelves
    -0.40
    CloseOperation
    -0.40
    AsUp
    -0.37
    posedge
    -0.37
    MIDDLEWARE
    -0.36
    マンド
    -0.35
    ſelf
    -0.35
    scriptcase
    -0.35
    POSITIVE LOGITS
     repatriation
    0.51
     wisest
    0.47
     Handling
    0.47
    益
    0.46
     another
    0.45
     Burning
    0.45
     retention
    0.44
    Handling
    0.44
     Retention
    0.44
    >|</
    0.44
    Activations Density 0.753%

    No Known Activations