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. 10-GEMMASCOPE-RES-16K
    6. 11845
    Prev
    Next
    INDEX
    Explanations

    terms related to tightness or tension in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Gand
    -0.70
     Brind
    -0.64
     gand
    -0.64
    [:,
    -0.63
    الحياه
    -0.63
    ogna
    -0.62
    Ol
    -0.62
    o
    -0.62
     Diener
    -0.61
    PostBack
    -0.61
    POSITIVE LOGITS
     tight
    1.65
     Tight
    1.58
    Tight
    1.53
    tight
    1.49
     tighter
    1.38
     tightness
    1.35
     tighten
    1.35
     tightening
    1.31
     tightened
    1.31
     tightly
    1.30
    Activations Density 0.218%

    No Known Activations