© 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. 10-RES_POST_32K-OAI
    11. 30023
    Prev
    Next
    INDEX
    Explanations

    repeated conjunctions in sentences

    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
    139
    +0.10
    0.5%
    254
    +0.10
    0.5%
    315
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.10
    0.07
    393
    +0.10
    0.07
    639
    +0.10
    0.05
    Negative Logits
    esthesia
    -0.90
    uria
    -0.83
    inian
    -0.74
    ciation
    -0.74
    cture
    -0.70
    urgical
    -0.70
    iasm
    -0.69
    ixture
    -0.69
    é¾įåĸļ士
    -0.66
    £ı
    -0.66
    POSITIVE LOGITS
     educators
    1.40
     poets
    1.39
     businessmen
    1.37
     journalists
    1.36
     entrepreneurs
    1.36
     thinkers
    1.36
     explorers
    1.36
     secretaries
    1.36
     technicians
    1.36
     administrators
    1.33
    Activations Density 0.764%

    No Known Activations