© 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. 10-RES_POST_32K-OAI
    11. 30732
    Prev
    Next
    INDEX
    Explanations

    possessive forms of nouns, particularly those indicating ownership or association

    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
    393
    +0.15
    0.7%
    15
    +0.13
    0.6%
    362
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.15
    0.08
    362
    +0.13
    0.08
    443
    +0.12
    0.04
    Negative Logits
    anguage
    -0.95
    lehem
    -0.90
    olkien
    -0.82
    ISO
    -0.79
    Downloadha
    -0.78
    hari
    -0.77
    ocument
    -0.77
    Reviewed
    -0.75
    assium
    -0.73
    Developer
    -0.73
    POSITIVE LOGITS
     wife
    1.14
    panic
    1.07
     own
    1.05
     grandson
    1.05
     granddaughter
    1.04
     widow
    1.03
     successor
    1.03
     girlfriend
    1.03
     eldest
    1.01
     penchant
    1.00
    Activations Density 0.513%

    No Known Activations