© 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. 29819
    Prev
    Next
    INDEX
    Explanations

    technical terms and structures related to programming and data access

    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
    520
    +0.16
    0.7%
    74
    +0.10
    0.4%
    424
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.16
    0.14
    520
    +0.10
    0.07
    137
    +0.10
    0.00
    Negative Logits
     Hein
    -0.85
    ãĥ¼ãĥĨãĤ£
    -0.79
    umpy
    -0.78
     Ruff
    -0.78
    soType
    -0.77
     Schneider
    -0.76
     grinning
    -0.71
    otto
    -0.71
     Merry
    -0.71
     knitting
    -0.70
    POSITIVE LOGITS
     Access
    2.08
    Access
    1.93
     access
    1.92
    access
    1.89
     accessibility
    1.60
     accessible
    1.50
     accessing
    1.48
     accessed
    1.39
    accessible
    1.34
     inaccessible
    1.32
    Activations Density 2.864%

    No Known Activations