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. 14858
    Prev
    Next
    INDEX
    Explanations

    situations involving past decisions and their consequences

    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
    setVerticalGroup
    -0.65
    AddTagHelper
    -0.53
    IUrlHelper
    -0.52
     виправивши
    -0.51
     fieldNum
    -0.50
    RenderAtEndOf
    -0.49
    StoryboardSegue
    -0.48
    IndentedString
    -0.48
    UserScript
    -0.46
    queryInterface
    -0.46
    POSITIVE LOGITS
     pessimism
    0.46
     hopelessness
    0.46
    ----</
    0.44
     feared
    0.42
    zbęd
    0.36
    annica
    0.36
     doubted
    0.35
     llorar
    0.35
     epä
    0.34
    LEGGI
    0.34
    Activations Density 0.418%

    No Known Activations