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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 15237
    Prev
    Next
    INDEX
    Explanations

    code snippets related to programming functions and method calls

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

    Code generation or programming-related terms

    np_acts-logits-general · gemini-2.0-flash

    MAKE_PROPERTY_D, mkern spaces

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    Personendaten
    -0.75
     betweenstory
    -0.74
    ConstraintMaker
    -0.65
    InSection
    -0.63
     UserModel
    -0.57
     tableName
    -0.53
    parsedMessage
    -0.52
    Tikang
    -0.50
    XmlAccessType
    -0.50
     namanya
    -0.48
    POSITIVE LOGITS
    parametrize
    0.49
     مُعرِّف
    0.49
     colorWith
    0.49
    glBind
    0.47
    libft
    0.47
     JADX
    0.46
     visual
    0.45
     partial
    0.45
     positive
    0.43
     onBind
    0.43
    Activations Density 0.295%

    No Known Activations