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. 14190
    Prev
    Next
    INDEX
    Explanations

    punctuations and non-alphanumeric characters

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

    emphasized asterisks

    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
     Pedersen
    -0.41
     victims
    -0.38
     Opfer
    -0.34
    ainte
    -0.33
    くて
    -0.33
     skin
    -0.33
     Rocha
    -0.32
     haughty
    -0.32
    0
    -0.32
    SpringRunner
    -0.32
    POSITIVE LOGITS
    存于互联网档案馆
    0.76
    parsedMessage
    0.67
    featureID
    0.67
    uxxxx
    0.63
     >=",
    0.60
    fromnode
    0.57
    gameserver
    0.55
    tvguidetime
    0.54
    ConstraintMaker
    0.52
     Bewußt
    0.52
    Activations Density 0.071%

    No Known Activations