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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 5694
    Prev
    Next
    INDEX
    Explanations

    dialogues involving character interactions and exchanges of information

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    dotenv
    -0.47
    devamını
    -0.45
    Numerade
    -0.44
    readFileSync
    -0.44
     ModelRenderer
    -0.43
     betweenstory
    -0.43
     itſelf
    -0.42
    FirstResponder
    -0.41
     createState
    -0.40
    fileInfo
    -0.40
    POSITIVE LOGITS
     ProtoMessage
    0.42
     تعدى
    0.41
     consultato
    0.40
     étrang
    0.40
     trataba
    0.39
     myself
    0.39
    IntoConstraints
    0.37
     дописавши
    0.37
    WindowConstants
    0.37
    …~
    0.36
    Activations Density 0.342%

    No Known Activations