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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 114329
    Prev
    Next
    INDEX
    Explanations

    terms related to software packages and system commands

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

    .pkg files, .stp files, .dtsi files, topics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    AndEndTag
    -1.06
     CreateTagHelper
    -1.02
    AddTagHelper
    -0.93
    ValueStyle
    -0.90
    DockStyle
    -0.89
     Chwiliwch
    -0.86
    TagMode
    -0.82
    RenderAtEndOf
    -0.82
     AssemblyCulture
    -0.82
     nahilalakip
    -0.81
    POSITIVE LOGITS
    相
    0.37
     Sim
    0.35
    SIM
    0.34
    .
    0.32
    ひ
    0.32
    </em>
    0.31
     Fle
    0.31
    com
    0.31
    py
    0.31
     file
    0.31
    Activations Density 0.051%

    No Known Activations