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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 35-GEMMASCOPE-RES-16K
    6. 7434
    Prev
    Next
    INDEX
    Explanations

    references to social commentary and critique, particularly about societal norms and attitudes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.35.hook_resid_post
    Hook Layer
    35
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    phosa
    -0.56
    vivencia
    -0.53
     statistique
    -0.44
     Maruti
    -0.44
     vété
    -0.43
    formin
    -0.43
     Miracles
    -0.43
    🇸
    -0.42
    :✨
    -0.42
    rlpool
    -0.41
    POSITIVE LOGITS
     defaultstate
    0.42
     bizarre
    0.41
    unken
    0.41
    abaret
    0.41
    :+:
    0.39
     Baudelaire
    0.39
     burlesque
    0.39
     Lichten
    0.39
     Industrial
    0.38
    ̈́
    0.38
    Activations Density 0.292%

    No Known Activations