© 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. 5755
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly quotes and apostrophes

    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
    280
    +0.11
    0.5%
    314
    +0.11
    0.5%
    666
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.11
    0.04
    442
    +0.11
    0.04
    160
    +0.10
    0.04
    Negative Logits
     Ezek
    -0.81
     Lumpur
    -0.80
     Qiao
    -0.76
     Ys
    -0.75
     Pieces
    -0.69
     transmissions
    -0.67
     Flavoring
    -0.66
     Univ
    -0.66
     Winds
    -0.65
     Tracking
    -0.65
    POSITIVE LOGITS
    athe
    0.86
    Jew
    0.85
    alcohol
    0.82
    environment
    0.80
    functional
    0.79
    aut
    0.79
    prem
    0.79
    piracy
    0.76
    trop
    0.76
    terrorist
    0.76
    Activations Density 0.184%

    No Known Activations