© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 2859
    Prev
    Next
    INDEX
    Explanations

    the word "no" and its variants, indicating negation or absence

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

    the phrase "have no" followed by a noun or concept.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/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.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    InstrumentedTest
    -0.41
    seende
    -0.40
     épo
    -0.39
     jsPsych
    -0.39
    underground
    -0.39
    species
    -0.38
    mes
    -0.38
    fabric
    -0.37
    les
    -0.37
    another
    -0.37
    POSITIVE LOGITS
     longer
    0.72
    цездатний
    0.65
    قایناقلار
    0.63
     affatto
    0.62
     particular
    0.62
    GEBURTSDATUM
    0.61
     other
    0.60
     discernible
    0.59
     entanto
    0.58
     apparent
    0.56
    Activations Density 0.157%

    No Known Activations