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 - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 85237
    Prev
    Next
    INDEX
    Explanations

    references to hexadecimal terminology and structures

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

    hex prefix

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    providedIn
    -0.54
    Pia
    -0.46
     Pia
    -0.42
    MIA
    -0.41
     pia
    -0.41
     Rockefeller
    -0.40
     PIA
    -0.40
     DBA
    -0.39
    cP
    -0.39
    pia
    -0.38
    POSITIVE LOGITS
     hex
    1.38
     Hex
    1.38
    Hex
    1.30
    hex
    1.23
     HEX
    1.20
    HEX
    1.07
    ToHex
    1.01
     hexagon
    1.00
     hexagonal
    0.95
    arangay
    0.93
    Activations Density 0.004%

    No Known Activations