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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 46019
    Prev
    Next
    INDEX
    Explanations

    code-related keywords, particularly those associated with exporting and requiring modules in programming context

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

    code keywords

    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
    aarrggbb
    -0.60
     ſind
    -0.60
     ſever
    -0.60
    GTCX
    -0.59
     ſaid
    -0.58
    MessageOf
    -0.57
     faſt
    -0.57
    <unused16>
    -0.56
    <unused8>
    -0.56
    [@BOS@]
    -0.56
    POSITIVE LOGITS
    .*")]
    0.53
     <<<<<<<<<<<<<<
    0.35
     Wikiseite
    0.35
    Panamoan
    0.33
     Vorstellung
    0.32
     Tremb
    0.32
     kentang
    0.31
     دیکھیے
    0.30
     NgModule
    0.30
     Erschein
    0.30
    Activations Density 0.001%

    No Known Activations