© 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. 8-RES_POST_32K-OAI
    11. 5099
    Prev
    Next
    INDEX
    Explanations

    instances of the word "reported" in various contexts related to information disclosure

    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
    201
    +0.09
    0.4%
    664
    +0.09
    0.4%
    441
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    195
    +0.09
    0.02
    488
    +0.09
    0.04
    441
    +0.09
    0.03
    Negative Logits
    abus
    -0.94
    ierre
    -0.80
    ivas
    -0.77
    atron
    -0.73
    quin
    -0.70
    essee
    -0.70
    inho
    -0.67
    ingo
    -0.66
    etry
    -0.66
    itus
    -0.65
    POSITIVE LOGITS
     sightings
    0.95
     sighting
    0.91
     seeing
    0.89
     successes
    0.85
     receiving
    0.84
     encountering
    0.82
     favorably
    0.80
     experiencing
    0.80
     success
    0.77
     feeling
    0.75
    Activations Density 0.321%

    No Known Activations