© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 46857
    Prev
    Next
    INDEX
    Explanations

    references to server responses and structural elements in a programming context

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

    web requests and code

    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
    GEBURTSDATUM
    -0.59
    RTCK
    -0.42
    पया
    -0.40
    Controllo
    -0.40
    addCriterion
    -0.38
    ioutil
    -0.36
    RuleContext
    -0.36
    Rüyada
    -0.35
    readObject
    -0.34
     aDecoder
    -0.34
    POSITIVE LOGITS
     jScrollPane
    0.51
    StoryboardSegue
    0.49
     /\.
    0.45
    CloseOperation
    0.45
    }`}>
    0.43
    ⎠
    0.43
    Filmografie
    0.42
    "](
    0.42
    Enllaços
    0.42
     NSCoder
    0.41
    Activations Density 0.007%

    No Known Activations