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 - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 69301
    Prev
    Next
    INDEX
    Explanations

    elements or properties related to programming or code structure

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

    wireshark, keyframes, inflate, exception

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    amante
    -0.49
    HasIndex
    -0.48
     Pris
    -0.46
     Princip
    -0.46
    Pris
    -0.46
    냅
    -0.45
     Rely
    -0.45
     Cuth
    -0.44
    Filters
    -0.44
     Incorpor
    -0.43
    POSITIVE LOGITS
    XmlAccessType
    0.53
    脚注の使い方
    0.52
    AccessorTable
    0.51
    ResourceManager
    0.50
    +#+#
    0.50
     useStyles
    0.49
    󠁢
    0.49
    figer
    0.47
    dafx
    0.45
    hyrchwyd
    0.45
    Activations Density 0.001%

    No Known Activations