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
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 368
    Prev
    Next
    INDEX
    Explanations

    references to mathematical concepts and vectors

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

    mathematical and scientific terms related to vectors, physics equations, and computational notation.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 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
     ब्रेकडाउन
    -0.45
     invokingState
    -0.38
     Mathf
    -0.36
    setcounter
    -0.33
     EconPapers
    -0.33
     BrowserModule
    -0.31
    UnusedPrivate
    -0.31
    tvguidetime
    -0.31
     vå
    -0.30
     relâche
    -0.30
    POSITIVE LOGITS
     vectors
    0.66
     vector
    0.61
     direction
    0.57
     vec
    0.56
     Vector
    0.56
    vectors
    0.56
     eigenvectors
    0.54
     Vectors
    0.54
     VECTOR
    0.54
    HomeAsUpEnabled
    0.54
    Activations Density 0.312%

    No Known Activations