© 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. 32051
    Prev
    Next
    INDEX
    Explanations

    assertions or claims related to findings and explanations in scientific research

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

    introduces relationships or consequences

    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
    ]');
    -0.55
    ()]);
    -0.55
     "");
    
    -0.52
    |};
    -0.52
    "])
    
    -0.50
    )";
    
    -0.50
    )])
    -0.49
    ?");
    -0.49
     '');
    -0.49
    )]);
    -0.49
    POSITIVE LOGITS
     ویکی‌پدی
    0.52
    angliski
    0.43
     noDo
    0.42
    setVerticalGroup
    0.42
    enterOuterAlt
    0.42
     tanong
    0.42
    
    0.41
    legungen
    0.40
    исленность
    0.39
     prevents
    0.39
    Activations Density 0.114%

    No Known Activations