© 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-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 2445
    Prev
    Next
    INDEX
    Explanations

    references to hypotheses and their validation in scientific proposals

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

    Word "hypothesis" or its variations

    np_acts-logits-general · gemini-2.0-flash

    hypothesis testing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    admir
    -0.52
    findpost
    -0.48
    gancia
    -0.48
     consulté
    -0.47
    intios
    -0.46
     '&#
    -0.45
    Portail
    -0.45
     atque
    -0.44
    ModelSerializer
    -0.44
    UnsafeEnabled
    -0.43
    POSITIVE LOGITS
     hypothesis
    1.95
     hypotheses
    1.77
     Hypothesis
    1.59
    hypothesis
    1.56
     hipótesis
    1.45
     Hypotheses
    1.43
    Hypothesis
    1.40
     hypothesized
    1.32
     hypothe
    1.30
     assumptions
    1.29
    Activations Density 0.835%

    No Known Activations