© 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. 8-RES_POST_32K-OAI
    11. 8051
    Prev
    Next
    INDEX
    Explanations

    the word "und" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    373
    +0.12
    0.5%
    459
    +0.11
    0.5%
    436
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.03
    187
    +0.11
    0.02
    699
    +0.09
    0.01
    Negative Logits
    ãĥīãĥ©ãĤ´ãĥ³
    -0.91
    utic
    -0.68
    fig
    -0.65
    ======
    -0.65
     toget
    -0.64
    kefeller
    -0.58
     Âł Âł Âł Âł
    -0.57
     gou
    -0.57
     adv
    -0.57
    grading
    -0.57
    POSITIVE LOGITS
    erest
    1.09
    erer
    1.04
    reth
    1.00
    rums
    0.99
    erers
    0.98
    ered
    0.96
    ecided
    0.96
    efined
    0.95
    oubtedly
    0.94
    lich
    0.92
    Activations Density 0.058%

    No Known Activations