© 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. 42118
    Prev
    Next
    INDEX
    Explanations

    phrases and terms related to analysis and transformations in data or processes

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

    analysis, parameters, opinion, efficiency

    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
     and
    -0.65
     however
    -0.54
    ,
    -0.52
     jednak
    -0.49
     però
    -0.47
     azonban
    -0.45
     kuitenkin
    -0.44
    however
    -0.44
     kuiten
    -0.43
    However
    -0.41
    POSITIVE LOGITS
    +#+#
    0.86
    SequentialGroup
    0.62
    UserScript
    0.57
    Datuak
    0.55
    évaluateur
    0.54
    msgSender
    0.53
    何より
    0.53
    principalTable
    0.53
     transférez
    0.52
    انجليز
    0.52
    Activations Density 1.196%

    No Known Activations