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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 5986
    Prev
    Next
    INDEX
    Explanations

    function declarations and the creation of new objects or arrays in programming code

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

    `function` and `Classes` declarations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pleaſure
    -0.69
     Houſe
    -0.66
     houſe
    -0.65
    AddTagHelper
    -0.63
     themſelves
    -0.60
     leaſt
    -0.60
     juſ
    -0.59
     ſche
    -0.58
     Anſ
    -0.58
     faſt
    -0.58
    POSITIVE LOGITS
    Cyfarwyddwr
    0.39
    NewGuid
    0.35
    MouseListener
    0.35
     kual
    0.35
    illige
    0.34
     uyar
    0.34
     diminta
    0.33
     minta
    0.33
     materialen
    0.32
    AsUp
    0.32
    Activations Density 0.062%

    No Known Activations