© 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. 18-GEMMASCOPE-RES-16K
    6. 9490
    Prev
    Next
    INDEX
    Explanations

    medical terms and references related to specific health conditions and treatments

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

    a temporal sequence

    np_acts-logits-general · gemini-2.0-flash

    before and after

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    awtextra
    -0.53
    __(/*!
    -0.53
    enetrable
    -0.50
     sonore
    -0.47
    trino
    -0.45
    lső
    -0.44
    }*/
    
    -0.43
    ungguhnya
    -0.43
    Tarih
    -0.42
    illigen
    -0.42
    POSITIVE LOGITS
     after
    2.59
     afterwards
    2.55
     afterward
    2.45
    after
    2.37
    AFTER
    2.21
     post
    2.16
     AFTER
    2.16
     после
    2.13
    After
    2.10
     після
    2.08
    Activations Density 1.336%

    No Known Activations