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-2-2B
    4. Residual Stream - 16k
    5. 16-GEMMASCOPE-RES-16K
    6. 5813
    Prev
    Next
    INDEX
    Explanations

    instances of reported injuries or incidents involving individuals in a workplace or educational setting

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

    appearing before verbs/actions

    np_acts-logits-general · gemini-2.0-flash

    hallway, company employees, grievance, supervisor

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 16-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    GEBURTSDATUM
    -0.96
     '\\;'
    -0.75
     utafitiHapana
    -0.62
     estimés
    -0.59
    енча
    -0.57
     setAddress
    -0.56
    SourceChecksum
    -0.55
    Linki
    -0.54
    ⓧ
    -0.54
    poran
    -0.53
    POSITIVE LOGITS
    NamedQueries
    0.56
     cafeteria
    0.53
     liderança
    0.52
     assigned
    0.52
    +#+#
    0.52
     pasillo
    0.51
     overheard
    0.51
     supervisor
    0.50
     ASSIGN
    0.49
    hasPermission
    0.49
    Activations Density 0.441%

    No Known Activations