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. 20-GEMMASCOPE-RES-16K
    6. 15687
    Prev
    Next
    INDEX
    Explanations

    terminology related to roles and effects in research contexts

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

    "role of" or "effect of"

    np_acts-logits-general · gemini-2.0-flash

    role or effect of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    DebuggerNonUser
    -0.94
     <<<<<<<<<<<<<<
    -0.74
     виправивши
    -0.72
     ſte
    -0.70
     myſelf
    -0.70
     stiefel
    -0.69
    ſelf
    -0.69
    ſelves
    -0.69
    goku
    -0.68
    UrlResolution
    -0.68
    POSITIVE LOGITS
     played
    0.56
     onBind
    0.50
     different
    0.48
     of
    0.48
     yks
    0.46
     dima
    0.44
    ‌آ
    0.44
    makeConstraints
    0.43
    marginBottom
    0.43
    CustomAttributes
    0.43
    Activations Density 0.438%

    No Known Activations