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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    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. 10775
    Prev
    Next
    INDEX
    Explanations

    legal definitions and stipulations regarding individual rights and qualifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     queſta
    -0.68
    sizeCache
    -0.66
     فريبيس
    -0.66
    TagMode
    -0.65
     Normdatei
    -0.64
     Мексичка
    -0.64
    BeginContext
    -0.60
     noqa
    -0.60
    OGND
    -0.59
    NameInMap
    -0.58
    POSITIVE LOGITS
     shall
    0.46
    AlterField
    0.45
     każde
    0.39
    hmigung
    0.38
     must
    0.38
     administrativos
    0.36
     Regel
    0.36
     procédure
    0.36
     procedure
    0.35
     laikā
    0.35
    Activations Density 0.062%

    No Known Activations