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-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 77161
    Prev
    Next
    INDEX
    Explanations

    The text primarily features casual discourse often involving user interactions, diverse topics ranging from personal anecdotes to technical discussions, expressed in an informal tone that encourages dialogue.

    eleuther_acts_top20 · gpt-4o-miniTriggered by @johnny

    instances of emphasis or contradiction in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     EnglishChoose
    -0.43
     kasarigan
    -0.41
     daß
    -0.40
    ViewStyle
    -0.38
     superiore
    -0.38
     Verbrechen
    -0.38
    BufferException
    -0.36
    wirkungen
    -0.36
     CURIAM
    -0.35
     []:
    -0.35
    POSITIVE LOGITS
    ConstraintMaker
    0.69
    出版年
    0.56
    (‘
    0.53
     ‘
    0.51
    …’
    0.51
    hoeddwyd
    0.49
    المناصب
    0.49
     (‘
    0.48
    ’
    0.48
     defaultstate
    0.47
    Activations Density 0.137%

    No Known Activations