© 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. 26928
    Prev
    Next
    INDEX
    Explanations

    repeated occurrences of the substring "ille"

    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
    710
    +0.15
    0.7%
    2
    +0.10
    0.4%
    583
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.15
    0.03
    99
    +0.10
    0.02
    288
    +0.10
    0.03
    Negative Logits
    anmar
    -0.74
    -+-+
    -0.74
    orate
    -0.73
    iating
    -0.72
    orage
    -0.69
    rush
    -0.68
    ounded
    -0.68
     DPR
    -0.67
    iates
    -0.67
    itars
    -0.67
    POSITIVE LOGITS
    ttes
    1.18
    tto
    1.15
    hammer
    1.01
    lla
    0.95
    tt
    0.94
    tta
    0.93
    tti
    0.89
    mma
    0.85
    ting
    0.84
    brand
    0.79
    Activations Density 0.269%

    No Known Activations