© 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-2-9B-IT
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 6629
    Prev
    Next
    INDEX
    Explanations

    references to looking or gaze, particularly involving eye contact

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

    looking or seeing

    np_acts-logits-general · gemini-2.0-flash

    eyes, gaze, looking

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     doInBackground
    -0.57
     actionMode
    -0.57
    Geplaatst
    -0.56
     informée
    -0.54
    +#+#
    -0.50
    ########.
    -0.49
     صوتيه
    -0.48
    úgó
    -0.48
    ביוגרפיה
    -0.47
    Tikang
    -0.47
    POSITIVE LOGITS
     eyes
    0.62
     glances
    0.61
     gaze
    0.57
     eye
    0.57
     glance
    0.56
     gazing
    0.55
     Eyes
    0.55
     EYES
    0.54
    Eyes
    0.54
    eyes
    0.53
    Activations Density 0.184%

    No Known Activations