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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 3840
    Prev
    Next
    INDEX
    Explanations

    mentions of the name "Jeff" and related epigenetic terminology

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

    Tokens after "Jeff" or "Ep"

    np_acts-logits-general · gemini-2.0-flash

    Jeff and ep followed by names or words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    __*/
    -0.87
    principalTable
    -0.83
    AddTagHelper
    -0.81
    
    -0.79
    RegisterType
    -0.73
    RegistryLite
    -0.69
    DeleteBehavior
    -0.63
    Toro
    -0.63
     MainAxisSize
    -0.63
    ConstraintMaker
    -0.63
    POSITIVE LOGITS
     Mez
    0.57
    ighed
    0.56
     Tep
    0.55
    EP
    0.53
     ep
    0.53
     EP
    0.52
    TypedDataSet
    0.50
    EPs
    0.49
    Mez
    0.48
     DataGridView
    0.47
    Activations Density 0.118%

    No Known Activations