Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15839
    Prev
    Next
    INDEX
    Explanations

    terms related to postal systems and mailing devices

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

    Postage or mail-related items

    np_acts-logits-general · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    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
     surla
    -0.65
     esternos
    -0.57
     onCreateView
    -0.46
     Audiodateien
    -0.45
    يميديا
    -0.42
    JspWriter
    -0.41
     Савезне
    -0.39
     Paglinawan
    -0.39
     مرئيه
    -0.38
    QMetaType
    -0.37
    POSITIVE LOGITS
     stamp
    1.36
     postal
    1.34
     stamps
    1.34
     Postal
    1.33
     Postmaster
    1.27
     USPS
    1.25
     Stamps
    1.24
     Stamp
    1.24
     Post
    1.23
     postage
    1.20
    Activations Density 0.069%

    No Known Activations