© 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 EvalsExportsinterp-engineNEWAPI 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. 28132
    Prev
    Next
    INDEX
    Explanations

    quotation marks and apostrophes indicating dialogue or speech

    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
    665
    +0.16
    0.7%
    49
    +0.14
    0.7%
    551
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    107
    +0.16
    0.06
    160
    +0.14
    0.06
    48
    +0.13
    0.04
    Negative Logits
     characteristic
    -0.76
     trailing
    -0.69
     glossy
    -0.67
     consequential
    -0.66
     cardinal
    -0.65
     prayers
    -0.64
     vague
    -0.64
     HUGE
    -0.63
     intrinsic
    -0.63
     occurring
    -0.63
    POSITIVE LOGITS
    create
    1.32
    make
    1.27
    take
    1.25
    bring
    1.23
    seek
    1.22
    keep
    1.22
    fill
    1.22
    come
    1.20
    grow
    1.19
    speak
    1.17
    Activations Density 0.386%

    No Known Activations