© 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. 30486
    Prev
    Next
    INDEX
    Explanations

    names and titles of individuals associated with events or activities

    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
    88
    +0.13
    0.6%
    548
    +0.10
    0.5%
    612
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.12
    232
    +0.10
    0.06
    314
    +0.10
    0.00
    Negative Logits
    _-
    -0.71
    ÄŁ
    -0.68
    ;;;;;;;;;;;;
    -0.67
    âĶģ
    -0.62
    ":[{"
    -0.62
     Aval
    -0.60
     balcon
    -0.60
    Ĺ
    -0.59
     PRESS
    -0.58
     spoiler
    -0.58
    POSITIVE LOGITS
    xon
    0.91
    heny
    0.70
    ockets
    0.69
    enhagen
    0.66
    zinski
    0.65
    nings
    0.64
    itiveness
    0.64
    orsche
    0.62
    rison
    0.61
    ulic
    0.60
    Activations Density 0.501%

    No Known Activations