© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 8928
    Prev
    Next
    INDEX
    Explanations

    phrases related to ongoing discussions or evaluations of various subjects

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

    Follows "is," "are," "was," "were," or "be."

    np_acts-logits-general · gemini-2.0-flash

    state or characteristic descriptions

    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
    γγ
    -0.56
     ostensibly
    -0.50
     glean
    -0.49
    |()
    -0.47
     toute
    -0.47
     occa
    -0.47
     afield
    -0.46
     Regal
    -0.46
     Zahl
    -0.46
     unwittingly
    -0.46
    POSITIVE LOGITS
     belonged
    0.90
     consisted
    0.87
     happened
    0.82
     belongs
    0.80
     belong
    0.79
    brainly
    0.75
     resulted
    0.74
     pinulongan
    0.74
     AssemblyCompany
    0.73
     existed
    0.73
    Activations Density 0.630%

    No Known Activations