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. 18-GEMMASCOPE-RES-16K
    6. 10815
    Prev
    Next
    INDEX
    Explanations

    instances of the word "when" indicating a temporal context or condition

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

    Code or equation fragments

    np_acts-logits-general · gemini-2.0-flash

    code comments and errors

    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
     Efq
    -1.03
     Theſe
    -0.86
    ſelf
    -0.85
     Paglinawan
    -0.77
     ſhould
    -0.76
     doubtnut
    -0.75
     moschino
    -0.75
     myſelf
    -0.73
     uſed
    -0.71
    ✨:
    -0.70
    POSITIVE LOGITS
     Chwiliwch
    0.69
    toBeTruthy
    0.61
      
    0.59
     */
    0.54
    enis
    0.53
     co
    0.52
    曖昧さ回避
    0.51
     OR
    0.50
    ٌ
    0.50
    </i>
    0.50
    Activations Density 0.245%

    No Known Activations