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. 31-GEMMASCOPE-RES-16K
    6. 15654
    Prev
    Next
    INDEX
    Explanations

    phrases that express concepts of trust and community engagement

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

    "trust" and its inflections

    np_acts-logits-general · gemini-2.0-flash

    trust

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    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
    -0.46
    ніципалі
    -0.45
    Carriera
    -0.44
    +#+#
    -0.43
    HasAnnotation
    -0.42
    zab
    -0.41
     actionPerformed
    -0.41
     actionMode
    -0.40
    spesies
    -0.39
    labus
    -0.39
    POSITIVE LOGITS
     trust
    3.33
    trust
    2.86
    Trust
    2.80
     Trust
    2.75
     TRUST
    2.69
     trusts
    2.56
     trusting
    2.42
    TRUST
    2.42
    信任
    2.23
     trusted
    2.20
    Activations Density 0.320%

    No Known Activations