Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 5-GEMMASCOPE-RES-16K
    6. 4890
    Prev
    Next
    INDEX
    Explanations

    terms related to "three" and "triad" structures 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 @ 5-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.5.hook_resid_post
    Hook Layer
    5
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    énéral
    -0.51
    ../../../
    -0.51
    ínsula
    -0.49
     Mank
    -0.47
    knecht
    -0.47
     remarquer
    -0.47
     twelfth
    -0.46
    läufe
    -0.46
    conio
    -0.46
    LAHOMA
    -0.46
    POSITIVE LOGITS
     trio
    1.19
     three
    1.08
     Three
    1.07
     triad
    1.04
    三
    1.04
     THREE
    1.03
     三
    1.03
    的三
    1.02
    三人
    1.01
     Trio
    1.01
    Activations Density 0.555%

    No Known Activations