© 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
    3. Gemma-2-2B
    5. 21-GEMMASCOPE-TRANSCODER-16K
    7. 5576
    Prev
    Next
    INDEX
    Explanations

    instances of the word "the" preceded by a 't' and followed by a verb.

    oai_token-act-pair · gemini-2.0-flash

    variations of the word 'much'

    np_token-act-pair-logits · gpt-4o-mini

    much, any, anything

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Occasionally
    -0.82
     Occasionally
    -0.75
     slightly
    -0.72
    بعض
    -0.72
     occasionally
    -0.71
     somewhat
    -0.70
    Slightly
    -0.70
     Slightly
    -0.70
    somewhat
    -0.69
     lidt
    -0.68
    POSITIVE LOGITS
     much
    1.28
     any
    1.12
    much
    1.09
     anything
    0.97
    Much
    0.94
     Much
    0.93
    any
    0.82
     muito
    0.82
     mucho
    0.80
     mycket
    0.79
    Activations Density 5.190%

    No Known Activations