© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 27825
    Prev
    Next
    INDEX
    Explanations

    references to basic human necessities and resources

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    143
    +0.11
    0.5%
    548
    +0.10
    0.5%
    50
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.11
    0.07
    393
    +0.10
    0.07
    546
    +0.10
    0.04
    Negative Logits
    ortium
    -0.75
    lishes
    -0.74
     Ic
    -0.71
    utenberg
    -0.69
    arna
    -0.68
     coerc
    -0.66
     fide
    -0.65
    uthor
    -0.65
     ABE
    -0.64
     Sabha
    -0.64
    POSITIVE LOGITS
     drink
    1.39
     Drink
    1.36
     drinks
    1.23
     beverages
    1.22
     beverage
    1.08
     thirsty
    1.07
     shelter
    1.00
     drinking
    0.97
     sleep
    0.95
     snacks
    0.94
    Activations Density 0.996%

    No Known Activations