© 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. 5622
    Prev
    Next
    INDEX
    Explanations

    the verb "have" in various contexts

    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
    653
    +0.12
    0.6%
    248
    +0.10
    0.5%
    380
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    605
    +0.12
    0.03
    7
    +0.10
    0.03
    566
    +0.10
    0.03
    Negative Logits
     uniformly
    -0.81
     overwhelmingly
    -0.77
     consistently
    -0.66
     invariably
    -0.65
    intosh
    -0.65
     continuously
    -0.64
    jriwal
    -0.61
     certific
    -0.60
    anding
    -0.59
     predominantly
    -0.58
    POSITIVE LOGITS
     someday
    0.82
     typo
    0.78
     inadvertently
    0.75
     misunder
    0.75
     underest
    0.72
    olla
    0.69
     accidentally
    0.68
     mistake
    0.68
     unintentionally
    0.66
    gotten
    0.64
    Activations Density 0.091%

    No Known Activations