© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 10-RES_POST_32K-OAI
    6. 29655
    Prev
    Next
    INDEX
    Explanations

    main themes related to transportation and significant events involving vehicles

    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
    407
    +0.11
    0.5%
    201
    +0.09
    0.4%
    574
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.11
    -0.01
    537
    +0.09
    0.01
    521
    +0.09
    0.00
    Negative Logits
    abase
    -0.73
    aeda
    -0.70
    ikan
    -0.69
    ciating
    -0.67
    anguage
    -0.65
    qqa
    -0.65
    ĸļ
    -0.65
    Ô
    -0.65
     ç¥ŀ
    -0.65
    estyles
    -0.64
    POSITIVE LOGITS
     collapsed
    0.90
     collapse
    0.76
     Racer
    0.70
     collided
    0.67
     collapses
    0.67
     perished
    0.66
     highway
    0.65
     survived
    0.61
     veh
    0.61
    cycl
    0.58
    Activations Density 0.041%

    No Known Activations