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. 20-GEMMASCOPE-RES-16K
    6. 13120
    Prev
    Next
    INDEX
    Explanations

    references to childhood memories and experiences

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

    Nostalgic reflections on early life

    np_acts-logits-general · gemini-2.0-flash

    early life memories

    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-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91
    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
    ablon
    -0.42
    modelBuilder
    -0.41
    createStatement
    -0.40
    pyplot
    -0.39
    AxisAlignment
    -0.39
    קישורים
    -0.39
     indisponible
    -0.39
     Appearance
    -0.38
    OpenHelper
    -0.38
     MessageBoxIcon
    -0.37
    POSITIVE LOGITS
     childhood
    0.71
     Childhood
    0.54
    Childhood
    0.53
     boyhood
    0.51
     Kindheit
    0.50
     memories
    0.50
    IsMutable
    0.47
     yıllar
    0.45
     infancia
    0.45
     years
    0.45
    Activations Density 0.229%

    No Known Activations