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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 10-RES-JB
    11. 2289
    Prev
    Next
    INDEX
    Explanations

    expressions of strong feelings or emotions, particularly passion

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.10.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    chanind/openwebtext-gemma
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    1,024
    Dataset
    ctigges/openwebtext-gemma-1024-cl
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.22
    1.2%
    896
    +0.14
    0.8%
    1828
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    896
    +0.22
    0.02
    806
    +0.14
    0.02
    1053
    +0.12
    0.02
    Negative Logits
    <bos>
    -2.72
    ⓧ
    -0.83
    /***
    -0.74
    ValueGenerated
    -0.71
    <?
    -0.71
    ///**
    -0.68
    webElementXpaths
    -0.67
    /*!
    -0.65
    /**
    -0.63
    AddColumn
    -0.63
    POSITIVE LOGITS
     pymysql
    1.25
     passion
    1.24
    Passion
    1.20
     Passion
    1.18
    passion
    1.16
     wien
    1.15
     passione
    1.14
     riva
    1.13
     aen
    1.08
     passions
    1.06
    Activations Density 0.033%

    No Known Activations