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. 8250
    Prev
    Next
    INDEX
    Explanations

    terms related to open access publishing and licensing

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

    Statements of no conflicts of interest

    np_acts-logits-general · gemini-2.0-flash

    author's declarations of interest

    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
     surla
    -0.76
    -------------</
    -0.62
     Anſ
    -0.56
     ujednoznacz
    -0.56
     itſelf
    -0.51
    verifyException
    -0.47
     ErrIntOverflow
    -0.47
     XCTest
    -0.46
    ientras
    -0.46
    InjectMocks
    -0.46
    POSITIVE LOGITS
    creativecommons
    0.75
     Attribution
    0.45
    lende
    0.40
     Creative
    0.38
     Numerade
    0.38
     الحره
    0.37
     freely
    0.36
    ORCID
    0.36
     صوتيه
    0.35
    Creative
    0.35
    Activations Density 0.060%

    No Known Activations