© 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. 8-RES_POST_32K-OAI
    11. 6282
    Prev
    Next
    INDEX
    Explanations

    proper nouns, specifically names of people and authors

    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
    +0.14
    0.7%
    218
    +0.14
    0.7%
    249
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.14
    0.07
    373
    +0.14
    -0.02
    232
    +0.11
    0.05
    Negative Logits
    ÄŁ
    -0.93
    ÅĤ
    -0.91
    Äį
    -0.88
    Äĩ
    -0.87
    eki
    -0.87
    merga
    -0.86
    ı
    -0.85
    pmwiki
    -0.85
     Bulgarian
    -0.82
     Kazakh
    -0.82
    POSITIVE LOGITS
     Reynolds
    0.91
     Elliott
    0.86
     Hancock
    0.85
     Lane
    0.84
     Collins
    0.84
     Allen
    0.83
     McKenzie
    0.81
     Baker
    0.81
     Watson
    0.81
     Evans
    0.81
    Activations Density 0.294%

    No Known Activations