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. 15-GEMMASCOPE-RES-16K
    6. 2753
    Prev
    Next
    INDEX
    Explanations

    issues related to responsibilities, compliance, and apologies in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/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.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    brtc
    -0.54
     disponibilités
    -0.54
     snippetHide
    -0.51
    abancı
    -0.51
    ANDUM
    -0.49
     głos
    -0.47
    üksek
    -0.47
    rines
    -0.46
     wings
    -0.45
    onInit
    -0.45
    POSITIVE LOGITS
     apology
    0.74
     reparation
    0.72
     compensation
    0.70
     reparations
    0.69
     retali
    0.69
     remedial
    0.67
     apologized
    0.67
     apologies
    0.66
     apologizing
    0.65
     risol
    0.64
    Activations Density 0.465%

    No Known Activations