Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 15121
    Prev
    Next
    INDEX
    Explanations

    terms and concepts associated with neural networks and artificial neural networks (ANNs)

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

    Cryptography or computer science terms

    np_acts-logits-general · gemini-2.0-flash

    public, private, key, training, binary

    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
     يتيمه
    -0.75
    .*")]
    -0.75
    клопе
    -0.71
     мәкал
    -0.69
     betweenstory
    -0.68
    +#+#
    -0.67
    AsUp
    -0.66
    transQ
    -0.66
    providedIn
    -0.64
    AndEndTag
    -0.63
    POSITIVE LOGITS
     anſ
    0.30
     ſeveral
    0.27
     seh
    0.27
     juſ
    0.26
     totalt
    0.26
    Verwaltung
    0.25
     chrétiens
    0.25
    Barrel
    0.25
     uſed
    0.24
     medži
    0.24
    Activations Density 0.642%

    No Known Activations