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. 15071
    Prev
    Next
    INDEX
    Explanations

    concepts related to performance, limitations, and biases in various systems or products

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

    that imply a negative consequence

    np_acts-logits-general · gemini-2.0-flash

    free from negative attributes

    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
    InitStruct
    -0.50
     متعلقه
    -0.50
     noDo
    -0.48
    MediaStore
    -0.44
    CloseOperation
    -0.44
    AutoScale
    -0.43
    Espèce
    -0.43
     esternos
    -0.42
    🇶
    -0.41
     المعيارى
    -0.41
    POSITIVE LOGITS
     untere
    0.47
     secretos
    0.47
     kuiten
    0.46
    SharedDtor
    0.45
     Tampoco
    0.45
     alcuna
    0.44
     ladrillo
    0.43
     aleación
    0.42
     vergüenza
    0.41
    VersionUID
    0.41
    Activations Density 0.344%

    No Known Activations