Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 131k
    5. 32-GEMMASCOPE-RES-131K
    6. 130008
    Prev
    Next
    INDEX
    Explanations

    references to task performance and cognitive engagement in experimental contexts

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

    level of difficulty

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.32.hook_resid_post
    Hook Layer
    32
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     المعيارى
    -0.67
     Paglinawan
    -0.61
    AndEndTag
    -0.57
    MethodManager
    -0.52
     يتيمه
    -0.50
    bootstrapcdn
    -0.50
     disambiguazione
    -0.49
     تضيفلها
    -0.47
     الحره
    -0.45
    PreInfinity
    -0.44
    POSITIVE LOGITS
     difficulty
    2.31
     difficult
    2.09
    difficulty
    2.00
     Difficulty
    1.92
     dificultad
    1.91
    difficult
    1.91
     Difficult
    1.86
    Difficulty
    1.86
     difficulté
    1.85
    Difficult
    1.84
    Activations Density 0.057%

    No Known Activations