© 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. 35-GEMMASCOPE-RES-16K
    6. 14421
    Prev
    Next
    INDEX
    Explanations

    specific terms associated with various objects and their descriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ientras
    -0.59
    ambién
    -0.54
    InitVars
    -0.53
     nakalista
    -0.52
    ()?;
    -0.52
     mêmes
    -0.49
    IBLIO
    -0.48
    AfterClass
    -0.48
    elfare
    -0.46
    amentlich
    -0.46
    POSITIVE LOGITS
     unlike
    0.89
    unlike
    0.82
    Unlike
    0.79
     traditionally
    0.77
     typically
    0.76
     Unlike
    0.75
     many
    0.74
    Typically
    0.73
     scientists
    0.69
     people
    0.69
    Activations Density 0.879%

    No Known Activations