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. 10580
    Prev
    Next
    INDEX
    Explanations

    the word "even" in various contexts

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

    "Even" followed by "though" or "if"

    np_acts-logits-general · gemini-2.0-flash

    even though

    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
     bahkan
    -0.78
    UnusedPrivate
    -0.75
    Either
    -0.71
     Either
    -0.71
    either
    -0.71
     nawet
    -0.70
    also
    -0.69
     навіть
    -0.65
    even
    -0.64
     даже
    -0.64
    POSITIVE LOGITS
     though
    1.84
    though
    1.48
     THOUGH
    1.43
     Though
    1.21
    Though
    1.21
     tho
    1.01
     worse
    0.89
    thou
    0.83
     thou
    0.80
    handed
    0.79
    Activations Density 0.079%

    No Known Activations