© 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. 6783
    Prev
    Next
    INDEX
    Explanations

    references to physical locations or specific places associated with actions or events

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

    Locations: floor, pool, street, bench, yard, ledge

    np_acts-logits-general · gemini-2.0-flash

    locations in text

    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
    ']):
    -0.70
    ]));
    
    -0.70
    ()");
    -0.69
    ()")
    -0.66
    ")));
    
    -0.64
    ")),
    -0.63
    ]))
    
    -0.63
    ())),
    -0.63
    "):
    
    -0.61
    '),
    
    -0.60
    POSITIVE LOGITS
     للمعارف
    0.90
    BeginInit
    0.69
     linkovi
    0.67
     viewDidLoad
    0.61
    vantages
    0.59
    ftagPool
    0.55
     insuffisamment
    0.55
     ExecuteAsync
    0.53
    ientôt
    0.52
    SuspendLayout
    0.51
    Activations Density 0.679%

    No Known Activations