© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 7108
    Prev
    Next
    INDEX
    Explanations

    phrases related to questioning or expressing disbelief

    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
    326
    +0.15
    0.7%
    530
    +0.14
    0.7%
    566
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.15
    0.11
    530
    +0.14
    0.11
    406
    +0.11
    0.09
    Negative Logits
    ghai
    -0.71
    olulu
    -0.69
    acity
    -0.68
    alys
    -0.68
    lahoma
    -0.68
    auga
    -0.65
    abase
    -0.63
    idae
    -0.63
    gio
    -0.63
    ks
    -0.62
    POSITIVE LOGITS
    ING
    1.81
    ER
    1.71
    ERS
    1.69
    ED
    1.66
    MENT
    1.65
    NESS
    1.65
    ITY
    1.64
    ION
    1.64
    ISM
    1.64
    WOR
    1.64
    Activations Density 2.415%

    No Known Activations