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-2-2B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 4536
    Prev
    Next
    INDEX
    Explanations

    adverbs that indicate assertiveness or carefulness in actions

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

    ending in "-ly" followed by a verb

    np_acts-logits-general · gemini-2.0-flash

    adverb + verb

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 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
     carefully
    -1.09
     ultimately
    -0.98
     efficiently
    -0.95
     properly
    -0.94
     easily
    -0.94
    caref
    -0.93
     essentially
    -0.92
     confidently
    -0.92
     reliably
    -0.91
     nicely
    -0.91
    POSITIVE LOGITS
    Попис
    0.63
    økt
    0.62
     and
    0.61
     ترین
    0.57
     сайті
    0.57
     "..\..\
    0.55
     pero
    0.55
    setof
    0.54
    且
    0.52
     оригіналу
    0.49
    Activations Density 0.350%

    No Known Activations