© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 4652
    Prev
    Next
    INDEX
    Explanations

    specific proper nouns or names related to individuals and entities

    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
    678
    +0.11
    0.5%
    275
    +0.10
    0.4%
    425
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.02
    232
    +0.10
    0.02
    373
    +0.09
    -0.01
    Negative Logits
    onday
    -0.62
    nit
    -0.59
    oshenko
    -0.58
     Highlights
    -0.58
     Pengu
    -0.58
     lob
    -0.57
    perse
    -0.57
    ipel
    -0.56
    dry
    -0.56
    mania
    -0.56
    POSITIVE LOGITS
    amiya
    0.82
     surn
    0.68
    isans
    0.64
    brand
    0.63
    enhagen
    0.63
     pronouns
    0.63
    onymous
    0.62
    onym
    0.61
    eland
    0.61
     besides
    0.59
    Activations Density 0.080%

    No Known Activations