© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 12949
    Prev
    Next
    INDEX
    Explanations

    words related to latitude and its variations

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

    The neuron activates for the letters "lat" at the beginning of words, particularly in proper names like "Latoya" and geographic/scientific terms like "latitudinal".

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    the word "Lat" followed by a number, typically as part of incomplete or fragmented words.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Portale
    -0.49
     חיצוניים
    -0.48
     []:
    -0.46
    hartt
    -0.41
    DockStyle
    -0.40
    henswürdigkeiten
    -0.40
     lyre
    -0.38
    iolo
    -0.38
    Marshall
    -0.38
     Niki
    -0.38
    POSITIVE LOGITS
     Lat
    0.79
     lat
    0.77
    Lat
    0.72
     LAT
    0.68
    LAT
    0.56
     latch
    0.55
    lat
    0.54
     latency
    0.51
     Latham
    0.51
     LATIN
    0.50
    Activations Density 0.132%

    No Known Activations