© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 9423
    Prev
    Next
    INDEX
    Explanations

    concepts related to improvement and effectiveness in various contexts

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

    improvement or progress

    np_acts-logits-general · gemini-2.0-flash

    greater and better outcomes

    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
     zijne
    -0.56
    nezeu
    -0.51
    eluarkan
    -0.47
     kuulu
    -0.47
    unggah
    -0.47
    rąg
    -0.46
     illustrationer
    -0.46
     παν
    -0.45
     långt
    -0.45
     voeten
    -0.45
    POSITIVE LOGITS
    Better
    0.70
     Better
    0.67
    better
    0.62
    easier
    0.61
     clearer
    0.60
    Easier
    0.60
     better
    0.59
     BETTER
    0.58
     easier
    0.58
     smoother
    0.57
    Activations Density 0.322%

    No Known Activations