© 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. 28109
    Prev
    Next
    INDEX
    Explanations

    punctuations and sentence structures that contribute to narrative flow

    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
    459
    +0.11
    0.5%
    378
    +0.10
    0.4%
    135
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    6
    +0.11
    0.26
    145
    +0.10
    0.20
    640
    +0.10
    0.18
    Negative Logits
    pter
    -0.85
     exponent
    -0.82
     paraph
    -0.77
     actu
    -0.77
     endors
    -0.76
     captcha
    -0.75
     surrog
    -0.75
     empath
    -0.74
     epigen
    -0.73
     predictive
    -0.73
    POSITIVE LOGITS
    Located
    1.73
     Located
    1.70
     Downtown
    1.36
    Near
    1.28
     downtown
    1.28
     nearby
    1.23
     Tours
    1.19
     Near
    1.15
    Lake
    1.15
    Residents
    1.15
    Activations Density 1.809%

    No Known Activations