© 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. 10-RES_POST_32K-OAI
    11. 29172
    Prev
    Next
    INDEX
    Explanations

    instances of the word "sign" and its variations, indicating a focus on references to signing or enrollment processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    104
    +0.12
    0.6%
    234
    +0.12
    0.5%
    679
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    104
    +0.12
    0.02
    679
    +0.12
    0.03
    314
    +0.11
    0.00
    Negative Logits
     Lerner
    -0.68
     nerv
    -0.68
    bian
    -0.65
    Bey
    -0.63
     Deng
    -0.63
     patience
    -0.62
     dismant
    -0.61
     sear
    -0.60
     bureaucracy
    -0.59
     Jiu
    -0.59
    POSITIVE LOGITS
    ificantly
    1.83
    ificant
    1.70
    ific
    1.58
    atures
    1.53
    ature
    1.51
    aling
    1.40
    aled
    1.28
    alling
    1.21
    als
    1.20
    IFIC
    1.16
    Activations Density 0.147%

    No Known Activations