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-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 77560
    Prev
    Next
    INDEX
    Explanations

    instances of the verb "be" indicating existence or state

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

    future states or actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     שוליים
    -0.42
     pungkasnya
    -0.40
    cartão
    -0.37
    Šaltiniai
    -0.37
     minOccurs
    -0.35
     dallo
    -0.34
     때문
    -0.33
     autorytatywna
    -0.33
    ⁱ
    -0.33
    但我
    -0.33
    POSITIVE LOGITS
     Jefus
    0.67
     Monfieur
    0.65
     Chriftian
    0.63
     poffe
    0.61
     raiſ
    0.60
     Diſ
    0.60
     ſever
    0.60
     Shakspeare
    0.60
     esternos
    0.60
     Eſ
    0.59
    Activations Density 0.004%

    No Known Activations