© 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. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 1147
    Prev
    Next
    INDEX
    Explanations

    references to the city of Memphis

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    1339
    +0.09
    0.3%
    50
    +0.07
    0.3%
    492
    +0.06
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1056
    +0.09
    0.03
    1691
    +0.07
    0.02
    1335
    +0.06
    0.02
    Negative Logits
    <?
    -0.77
    /**
    -0.75
    <bos>
    -0.69
    
    
    -0.67
    public
    -0.64
    /*
    -0.64
    ുറ
    -0.60
    Xuất
    -0.57
    private
    -0.56
    protected
    -0.56
    POSITIVE LOGITS
     Minang
    1.30
     accla
    1.29
     maneu
    1.22
     lele
    1.15
     intermitt
    1.14
     Memphis
    1.12
     jaya
    1.12
     dises
    1.11
     reluct
    1.11
     emphat
    1.10
    Activations Density 0.142%

    No Known Activations