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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 14458
    Prev
    Next
    INDEX
    Explanations

    different forms of the verb "run."

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

    Forms of "run"

    np_acts-logits-general · gemini-2.0-flash

    run followed by prepositions

    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
    enumi
    -0.78
    Géographie
    -0.68
     purpoſe
    -0.67
     doubtnut
    -0.65
    theless
    -0.65
     ſtate
    -0.64
     AssemblyTitle
    -0.64
     myſelf
    -0.61
    ⓘ
    -0.60
    الإنجليزية
    -0.60
    POSITIVE LOGITS
     run
    1.22
     Run
    1.21
     runs
    1.18
     Running
    1.16
     ran
    1.15
     running
    1.15
    running
    1.14
     Runs
    1.14
    Run
    1.13
     RUN
    1.13
    Activations Density 0.174%

    No Known Activations