© 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-2-2B
    4. Residual Stream - 16k
    5. 24-GEMMASCOPE-RES-16K
    6. 3803
    Prev
    Next
    INDEX
    Explanations

    negative prefixes and terms related to lack or deficiency

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

    The token "und"

    np_acts-logits-general · gemini-2.0-flash

    words starting with und

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    сылкі
    -0.56
    textTheme
    -0.56
    parsedMessage
    -0.55
    WriteAttribute
    -0.51
     pleaſure
    -0.47
    Rujuakan
    -0.47
    Pautan
    -0.47
    ReferenceEquals
    -0.46
     Chapin
    -0.46
    sile
    -0.46
    POSITIVE LOGITS
     AssemblyTitle
    0.87
     CreateTagHelper
    0.84
     surla
    0.61
    thodox
    0.58
     дописавши
    0.57
    ERVIS
    0.53
    ODB
    0.53
    󠁢
    0.52
    اوية
    0.51
    __':
    0.51
    Activations Density 0.171%

    No Known Activations