© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 28476
    Prev
    Next
    INDEX
    Explanations

    aspects related to education, experience, and professional qualifications

    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
    361
    +0.13
    0.6%
    95
    +0.10
    0.4%
    633
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.13
    651
    +0.10
    0.10
    15
    +0.10
    0.07
    Negative Logits
    omever
    -0.75
    ãĥ¤
    -0.73
     Ones
    -0.68
     Carbuncle
    -0.65
    inki
    -0.63
     Cosponsors
    -0.63
     binaries
    -0.62
     particular
    -0.61
    "]=>
    -0.61
     Thumbnails
    -0.60
    POSITIVE LOGITS
     captivity
    0.91
     spent
    0.88
     uninterrupted
    0.88
     devoted
    0.84
    inement
    0.83
     elapsed
    0.83
     culminating
    0.82
    interrupted
    0.80
     culminated
    0.76
     neglect
    0.75
    Activations Density 1.740%

    No Known Activations