© 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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 99763
    Prev
    Next
    INDEX
    Explanations

    references to various categories or groups of things or individuals

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

    listing subsequent items

    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
     للمعارف
    -0.48
    exitRule
    -0.48
    ConstraintMaker
    -0.47
    olverine
    -0.47
    rungsseite
    -0.47
     perſon
    -0.47
    ፩
    -0.44
    sizeCache
    -0.44
    AnimationsModule
    -0.44
     EdgeInsets
    -0.44
    POSITIVE LOGITS
    Spoljašnje
    0.49
    +#+#
    0.45
     OFDb
    0.44
    MessageTagHelper
    0.43
    RTGC
    0.39
     dotyczą
    0.36
    ----</
    0.35
    ècie
    0.35
     stuff
    0.35
     compito
    0.35
    Activations Density 0.029%

    No Known Activations