© 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. 28733
    Prev
    Next
    INDEX
    Explanations

    elements and functions related to programming or coding structures

    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
    711
    +0.12
    0.6%
    429
    +0.12
    0.5%
    58
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    134
    +0.12
    0.07
    429
    +0.12
    0.05
    393
    +0.11
    0.12
    Negative Logits
    omen
    -0.88
     Constantine
    -0.78
    wat
    -0.77
    cham
    -0.75
     thunder
    -0.73
    wich
    -0.72
     plum
    -0.72
    adow
    -0.72
    lot
    -0.71
    jah
    -0.71
    POSITIVE LOGITS
     Self
    2.25
     self
    2.20
    Self
    2.10
    self
    2.06
     selves
    1.64
    selves
    1.38
    ELF
    1.23
     autonomous
    1.17
     autonomy
    1.14
     selfies
    1.14
    Activations Density 3.008%

    No Known Activations