© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 30044
    Prev
    Next
    INDEX
    Explanations

    occurrences of the name "Jim."

    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
    218
    +0.13
    0.6%
    430
    +0.13
    0.6%
    392
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.02
    141
    +0.13
    0.02
    232
    +0.11
    0.02
    Negative Logits
     selective
    -0.76
     subsequ
    -0.68
    Constructed
    -0.67
    lihood
    -0.66
     Borough
    -0.65
     princ
    -0.59
     screened
    -0.57
     livest
    -0.56
    BOOK
    -0.56
     stitch
    -0.56
    POSITIVE LOGITS
    mie
    1.55
    enez
    1.54
    bo
    1.20
    iny
    1.17
    inez
    1.06
    mys
    1.02
    mi
    0.97
    bos
    0.96
    my
    0.90
    upiter
    0.89
    Activations Density 0.050%

    No Known Activations