© 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
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15498
    Prev
    Next
    INDEX
    Explanations

    assertions about faith and belief systems

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

    expressions of personal belief, conviction, or confidence in religious or interpersonal contexts.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    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
     ब्रेकडाउन
    -0.43
     lizenzfreies
    -0.40
    phrine
    -0.40
    }{*}{}
    -0.39
    ContextMenu
    -0.38
     understatement
    -0.37
     Zlat
    -0.36
    Reis
    -0.36
    kward
    -0.35
     repar
    -0.35
    POSITIVE LOGITS
     belief
    2.27
     faith
    2.08
     Belief
    1.98
     believe
    1.98
    belief
    1.92
    Belief
    1.88
     believing
    1.86
    believe
    1.82
     beliefs
    1.81
     Believe
    1.77
    Activations Density 0.371%

    No Known Activations