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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 30-GEMMASCOPE-RES-16K
    6. 12763
    Prev
    Next
    INDEX
    Explanations

    coding logic related to data structure validations and processing

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

    Code checking for empty fields/values

    np_acts-logits-general · gemini-2.0-flash

    checking if lists are empty

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 30-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_30/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.30.hook_resid_post
    Hook Layer
    30
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     "..\..\..\
    -0.38
     strtok
    -0.36
     CGPoint
    -0.35
     PopupWindow
    -0.35
    tagHelperRunner
    -0.34
    どちら
    -0.32
    organ
    -0.32
    getenv
    -0.32
     SwitchCompat
    -0.31
    FXMLLoader
    -0.31
    POSITIVE LOGITS
     ujednoznacz
    0.75
     cherchés
    0.67
    Tikang
    0.63
    Jîn
    0.61
    tagPool
    0.61
     ISNI
    0.60
    Diweddarwch
    0.60
    fjspx
    0.60
     հղումներ
    0.59
     varandra
    0.58
    Activations Density 1.165%

    No Known Activations