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. 10151
    Prev
    Next
    INDEX
    Explanations

    specific technical terms and methodologies related to separation techniques and processes

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

    A technique or process is being described

    np_acts-logits-general · gemini-2.0-flash

    describes a process or technique

    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
    IActionResult
    -0.61
    ensity
    -0.52
    RSpec
    -0.49
    Dichloropropane
    -0.47
     العنوان
    -0.47
     otomatig
    -0.46
    ものが
    -0.45
     imageNamed
    -0.45
     üzerinde
    -0.45
    ureusement
    -0.44
    POSITIVE LOGITS
     polega
    1.56
     involves
    1.54
     involve
    1.38
    volves
    1.21
     заключается
    1.15
     INVOL
    1.10
     consiste
    1.10
     involving
    1.09
     whereby
    1.07
     waarbij
    1.02
    Activations Density 1.060%

    No Known Activations