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. 22-GEMMASCOPE-RES-16K
    6. 13689
    Prev
    Next
    INDEX
    Explanations

    terms related to apologies and expressions of remorse

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

    apologies and remorse

    np_acts-logits-general · gemini-2.0-flash

    apology and remorse

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     whim
    -0.48
    statusText
    -0.45
    ğından
    -0.45
     Auff
    -0.44
     prak
    -0.44
     suspiciously
    -0.42
     suspicious
    -0.42
     нат
    -0.41
    cestershire
    -0.41
     sceptical
    -0.40
    POSITIVE LOGITS
     apology
    2.70
     apologized
    2.52
     apologize
    2.52
     apologizing
    2.48
     apologies
    2.43
     apologise
    2.32
     apologised
    2.29
     apolog
    2.17
     apologe
    1.89
     sorry
    1.84
    Activations Density 0.399%

    No Known Activations