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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 24-GEMMASCOPE-RES-16K
    11. 4910
    Prev
    Next
    INDEX
    Explanations

    instances of the word "think" and its variations indicating critical and thoughtful engagement

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

    "think" followed by "about" or a preposition

    np_acts-logits-general · gemini-2.0-flash

    think about

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     législ
    -0.56
    uskas
    -0.53
     ögon
    -0.52
    puestas
    -0.51
    umumkan
    -0.51
     espoir
    -0.51
     restantes
    -0.49
     untere
    -0.48
     votes
    -0.48
     votación
    -0.48
    POSITIVE LOGITS
     about
    0.99
     twice
    0.89
     aloud
    0.84
     critically
    0.80
     deeply
    0.78
     logically
    0.77
    tank
    0.75
     ahead
    0.75
     hard
    0.74
     differently
    0.74
    Activations Density 0.072%

    No Known Activations