© 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. 31650
    Prev
    Next
    INDEX
    Explanations

    instances of the name "Maria" and its variations, indicating a focus on references to a specific person

    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
    257
    +0.10
    0.4%
    484
    +0.10
    0.4%
    444
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.03
    374
    +0.10
    0.00
    612
    +0.10
    0.02
    Negative Logits
    urities
    -0.68
    merce
    -0.68
    GBT
    -0.68
    onomy
    -0.67
    ornia
    -0.67
    neys
    -0.66
    orph
    -0.66
    ridges
    -0.65
    fty
    -0.64
    INTON
    -0.64
    POSITIVE LOGITS
     Theresa
    0.98
     Teresa
    0.90
    ppa
    0.82
     Elena
    0.80
    Maria
    0.71
     Zak
    0.69
     Isabel
    0.69
    ICA
    0.68
    eor
    0.66
     Bam
    0.65
    Activations Density 0.230%

    No Known Activations