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. 15472
    Prev
    Next
    INDEX
    Explanations

    information related to school policies and student academic requirements

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

    Follows numbers, "the", or start of turn

    np_acts-logits-general · gemini-2.0-flash

    titles and positions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     лек
    -0.49
     UNIVERS
    -0.44
     university
    -0.43
     universities
    -0.42
     pedagogical
    -0.40
     UNIVERSITY
    -0.40
    university
    -0.39
     lecturer
    -0.38
     pedagog
    -0.38
     schoolchildren
    -0.38
    POSITIVE LOGITS
    AndEndTag
    0.66
     CreateTagHelper
    0.63
    ñora
    0.60
     Señora
    0.60
     typelib
    0.59
     resourceCulture
    0.58
     Mrs
    0.58
     Yearbook
    0.57
    Mrs
    0.56
    parsedMessage
    0.54
    Activations Density 0.133%

    No Known Activations