© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 106911
    Prev
    Next
    INDEX
    Explanations

    expressions related to difficulty and challenges encountered in various tasks

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

    difficulty and challenge

    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
    IsMutable
    -0.44
    ($__
    -0.43
     Paglinawan
    -0.42
    EditorBrowsable
    -0.40
    Aktualisiert
    -0.38
     stanovnika
    -0.38
     numberWith
    -0.37
    
    -0.37
     吋
    -0.36
    WebServlet
    -0.36
    POSITIVE LOGITS
    Difficult
    0.69
     difficult
    0.68
     Difficult
    0.67
    difficult
    0.67
     difficulty
    0.62
    Hard
    0.60
    hard
    0.59
     difficulties
    0.58
     dificultad
    0.57
    Difficulty
    0.56
    Activations Density 0.063%

    No Known Activations