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. 91446
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns and possessive forms of nouns

    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
     chrétien
    -0.50
     Dingen
    -0.49
    jalá
    -0.48
     feroit
    -0.46
     Dinge
    -0.42
     auroit
    -0.41
     Könige
    -0.40
     pieniądze
    -0.40
     delà
    -0.40
     things
    -0.39
    POSITIVE LOGITS
     own
    1.02
     ihrer
    0.82
     seiner
    0.80
     his
    0.79
     their
    0.79
     svoj
    0.79
     свой
    0.79
     ihres
    0.78
     His
    0.78
     своя
    0.78
    Activations Density 0.001%

    No Known Activations