© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 10-RES-JB
    11. 3211
    Prev
    Next
    INDEX
    Explanations

    instances of viewer feedback and critique

    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
    +0.21
    0.8%
    1343
    +0.11
    0.4%
    381
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1150
    +0.21
    0.04
    964
    +0.11
    0.05
    1445
    +0.11
    0.06
    Negative Logits
    <bos>
    -2.39
    ⓧ
    -0.93
    <?
    -0.83
    /***
    -0.82
    
    -0.77
    /**
    -0.77
    /*
    -0.65
     impegno
    -0.65
    ValueGenerated
    -0.62
    <?
    -0.60
    POSITIVE LOGITS
     indescri
    0.83
     juf
    0.76
     unspeak
    0.75
     ditto
    0.73
     toti
    0.70
     maneu
    0.67
     vne
    0.66
     pylab
    0.66
     dă
    0.64
     quoque
    0.64
    Activations Density 0.516%

    No Known Activations