Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 16-GEMMASCOPE-RES-65K
    6. 53196
    Prev
    Next
    INDEX
    Explanations

    expressions of regret or remorse

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

    regret about or not

    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-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    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
    fjspx
    -0.80
    divarius
    -0.68
     Shand
    -0.57
    AndroidJUnit
    -0.54
    Oliver
    -0.53
    IUrlHelper
    -0.53
    AccessorTable
    -0.52
    ClientSize
    -0.52
    AntiForgeryToken
    -0.51
    quiao
    -0.51
    POSITIVE LOGITS
    tably
    1.28
     Regret
    1.09
     regret
    1.05
    Regret
    0.91
     regretted
    0.88
     regrets
    0.87
    ful
    0.80
     regrettable
    0.73
     pédagogique
    0.71
    fully
    0.70
    Activations Density 0.016%

    No Known Activations