© 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. 30178
    Prev
    Next
    INDEX
    Explanations

    the presence of legal or regulatory terms related to accountability and enforcement

    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
    125
    +0.12
    0.6%
    577
    +0.10
    0.5%
    265
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.12
    0.13
    539
    +0.10
    0.12
    143
    +0.10
    0.12
    Negative Logits
    soDeliveryDate
    -0.75
    aye
    -0.71
    ipedia
    -0.66
    cream
    -0.65
    thood
    -0.64
     Birthday
    -0.61
    spe
    -0.61
    RAFT
    -0.61
    deck
    -0.59
    Champ
    -0.59
    POSITIVE LOGITS
    "(
    0.91
    ''.
    0.81
    Added
    0.75
    (
    0.73
    :(
    0.70
    essee
    0.69
     duly
    0.67
     conspic
    0.66
    (&
    0.63
     ``(
    0.63
    Activations Density 1.408%

    No Known Activations