© 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. 8-RES_POST_32K-OAI
    6. 5875
    Prev
    Next
    INDEX
    Explanations

    mentions of a franchise in a sports or entertainment context

    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
    583
    +0.15
    0.7%
    288
    +0.14
    0.7%
    566
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.15
    0.02
    652
    +0.14
    0.01
    205
    +0.12
    0.01
    Negative Logits
    board
    -0.84
    boards
    -0.78
    ynthesis
    -0.74
     Marketable
    -0.70
     Thou
    -0.70
    rough
    -0.66
     nerv
    -0.66
    inx
    -0.65
    mith
    -0.64
    Downloadha
    -0.64
    POSITIVE LOGITS
    eers
    0.99
    rats
    0.78
    arily
    0.74
    Wars
    0.74
    PET
    0.73
    ally
    0.73
    e
    0.72
     cornerstone
    0.70
    pps
    0.70
    esan
    0.69
    Activations Density 0.095%

    No Known Activations