© 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. 28-GEMMASCOPE-RES-16K
    6. 8707
    Prev
    Next
    INDEX
    Explanations

    technical requirements and specifications in documents

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

    Prepositions and conjunctions following verbs, adjectives, or nouns

    np_acts-logits-general · gemini-2.0-flash

    precedes phrases like 'by', 'shape', 'well', 'body', 'the', 'be', 'on', 'in', '.'

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.28.hook_resid_post
    Hook Layer
    28
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hogy
    -0.38
     Italijanski
    -0.35
     jotka
    -0.35
    audiovisuel
    -0.31
    はじめに
    -0.31
    rodek
    -0.31
     którzy
    -0.30
    kään
    -0.30
     pagină
    -0.30
    oarece
    -0.30
    POSITIVE LOGITS
     wireType
    0.70
     चीज़ों
    0.53
    setVerticalGroup
    0.50
    ArgsConstructor
    0.49
    InteropServices
    0.47
     MainAxisSize
    0.47
     geworden
    0.47
    OCCURRED
    0.46
    awtextra
    0.46
     autorytatywna
    0.44
    Activations Density 1.436%

    No Known Activations