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. 10750
    Prev
    Next
    INDEX
    Explanations

    common phrases and high-frequency function words in the text

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

    relative clauses

    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
     pinulongan
    -0.68
    +#+
    -0.61
     beginnetje
    -0.60
     myſelf
    -0.57
     незавершена
    -0.57
     Autorisations
    -0.57
    ſelves
    -0.57
    الحياه
    -0.56
    expandindo
    -0.56
     $_"
    -0.54
    POSITIVE LOGITS
    principalColumn
    0.40
    StartTag
    0.40
    cloudfront
    0.35
    verteilung
    0.35
    навли
    0.35
    AsUp
    0.35
    riwal
    0.34
    大
    0.31
     econô
    0.30
     cr
    0.29
    Activations Density 0.028%

    No Known Activations