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. 3839
    Prev
    Next
    INDEX
    Explanations

    references to specific types and classifications of medical conditions and their commonality

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

    Follows "type," "form," or "types"

    np_acts-logits-general · gemini-2.0-flash

    types and kinds

    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
     betweenstory
    -0.77
     ModelExpression
    -0.52
    mtx
    -0.51
    ători
    -0.46
    writeFieldEnd
    -0.46
    IntoConstraints
    -0.46
    ConstraintMaker
    -0.46
    Tikang
    -0.44
     AssemblyTitle
    -0.44
    الإنجليزية
    -0.44
    POSITIVE LOGITS
     types
    2.15
     type
    2.08
    types
    1.75
     kinds
    1.75
    type
    1.74
     TYPES
    1.71
     tipo
    1.71
     Types
    1.70
    Types
    1.64
     kind
    1.63
    Activations Density 0.629%

    No Known Activations