© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 4315
    Prev
    Next
    INDEX
    Explanations

    instances of the word "regard" and its variations, indicating discussions about perspectives or opinions

    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
    314
    +0.13
    0.6%
    163
    +0.10
    0.4%
    624
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    386
    +0.13
    0.01
    624
    +0.10
    0.01
    253
    +0.09
    0.01
    Negative Logits
    mith
    -0.69
     Sieg
    -0.65
    wagen
    -0.64
     brainstorm
    -0.61
     Greenberg
    -0.59
     uphill
    -0.59
    backer
    -0.56
     Lans
    -0.56
    zzy
    -0.54
    ammy
    -0.54
    POSITIVE LOGITS
    ibly
    0.92
    ible
    0.91
    èĥ
    0.91
    phas
    0.88
    enance
    0.86
    ively
    0.80
    代
    0.79
    ä
    0.75
    åº
    0.75
    perature
    0.75
    Activations Density 0.041%

    No Known Activations