© 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-2-9B
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 14063
    Prev
    Next
    INDEX
    Explanations

    patterns related to programming or coding structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Autoritní
    -0.61
    цездатний
    -0.57
    ukone
    -0.54
    Vidite
    -0.52
    ativement
    -0.52
    wapV
    -0.51
    RTLU
    -0.50
    __(/*!
    -0.49
    ValueStyle
    -0.48
    الحياه
    -0.48
    POSITIVE LOGITS
     máscaras
    0.49
     sisält
    0.46
     imprimible
    0.44
     rę
    0.44
     BitSet
    0.44
     abstrait
    0.43
     içeri
    0.43
     mask
    0.43
    bitset
    0.41
     transparencia
    0.40
    Activations Density 1.595%

    No Known Activations