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-2-2B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 10727
    Prev
    Next
    INDEX
    Explanations

    references to information systems and data transfer

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

    preceding prepositions

    np_acts-logits-general · gemini-2.0-flash

    methods or circumstances

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
    msgTypes
    -0.90
     Савезне
    -0.81
    esterday
    -0.77
     SwitchCompat
    -0.77
     насељу
    -0.76
    DoubleQuotes
    -0.74
     оригіналу
    -0.74
     الاطلاع
    -0.72
    hugh
    -0.72
     nakalista
    -0.71
    POSITIVE LOGITS
     efficiently
    0.82
     via
    0.76
     manually
    0.75
     faster
    0.74
     secara
    0.74
     remotely
    0.71
     privately
    0.70
     without
    0.70
     responsibly
    0.69
     using
    0.68
    Activations Density 1.077%

    No Known Activations