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. 22-GEMMASCOPE-RES-16K
    6. 5638
    Prev
    Next
    INDEX
    Explanations

    references to geographical locations and their characteristics

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

    Follows punctuation or statement

    np_acts-logits-general · gemini-2.0-flash

    statement not hearsay

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     itſelf
    -2.04
     myſelf
    -2.00
     themſelves
    -1.83
     Jefus
    -1.70
     Efq
    -1.67
     Monfieur
    -1.64
     himſelf
    -1.61
    ^(@)
    -1.57
     purpoſe
    -1.55
     houſe
    -1.54
    POSITIVE LOGITS
     il
    0.84
    sem
    0.84
    lo
    0.84
     der
    0.83
    tr
    0.82
    una
    0.81
    Ã
    0.80
    kla
    0.80
    0.80
    rag
    0.80
    Activations Density 2.079%

    No Known Activations