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. 24-GEMMASCOPE-RES-16K
    6. 1065
    Prev
    Next
    INDEX
    Explanations

    connective words or conjunctions in text

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

    "and" (possibly questions)

    np_acts-logits-general · gemini-2.0-flash

    and or

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    RegressionTest
    -0.74
    GEBURTSDATUM
    -0.64
     disambiguazione
    -0.63
    Personensuche
    -0.57
     Wikimedijinoj
    -0.51
    PreferredItem
    -0.48
    DIPSETTING
    -0.47
    mtrl
    -0.45
    tagHelperRunner
    -0.45
    +#+#
    -0.45
    POSITIVE LOGITS
    
    0.45
     einstellen
    0.39
    Legături
    0.38
    drücken
    0.37
    źródło
    0.36
     interesantes
    0.36
    levance
    0.35
    kozá
    0.34
     Bedien
    0.34
     možnosti
    0.34
    Activations Density 0.544%

    No Known Activations