© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 2-GEMMASCOPE-RES-16K
    6. 13491
    Prev
    Next
    INDEX
    Explanations

    phrases indicating prevalence or commonality among subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     يتيمه
    -0.49
    oa̍t
    -0.40
    Painter
    -0.38
     evacuated
    -0.38
    choly
    -0.37
    RetentionPolicy
    -0.37
     pikir
    -0.36
    ksikon
    -0.36
     spontan
    -0.36
    chemise
    -0.36
    POSITIVE LOGITS
     hits
    1.02
     hitting
    0.91
    hits
    0.89
     Hits
    0.89
     majority
    0.79
    Hits
    0.69
    hitting
    0.68
    majority
    0.68
     hitters
    0.59
     batting
    0.54
    Activations Density 0.096%

    No Known Activations