© 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. 38-GEMMASCOPE-RES-16K
    6. 7992
    Prev
    Next
    INDEX
    Explanations

    instances of the word "push" and its variations in context, often related to action and progression

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    jsonwebtoken
    -0.36
    ushort
    -0.31
    ary
    -0.31
     bäst
    -0.31
    wpdb
    -0.30
    ISE
    -0.30
    nameof
    -0.30
     nadzie
    -0.29
    ations
    -0.29
    umis
    -0.29
    POSITIVE LOGITS
     pushed
    0.84
     aside
    0.79
    pushed
    0.79
     pushes
    0.75
     shove
    0.74
     PUSH
    0.72
     Pushing
    0.72
     Push
    0.72
     shoved
    0.71
     push
    0.70
    Activations Density 0.169%

    No Known Activations