© 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. 31808
    Prev
    Next
    INDEX
    Explanations

    verbs related to the act of creating or producing something

    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
    373
    +0.15
    0.7%
    741
    +0.10
    0.4%
    177
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    402
    +0.15
    0.10
    366
    +0.10
    0.05
    570
    +0.10
    0.03
    Negative Logits
     Retrieved
    -0.69
    èª
    -0.67
    stood
    -0.60
    reply
    -0.59
    hips
    -0.59
    ingen
    -0.59
    ahu
    -0.58
    DEF
    -0.58
     keynote
    -0.56
    bourg
    -0.56
    POSITIVE LOGITS
     sure
    1.43
     Sure
    0.88
     note
    0.81
     certain
    0.80
    sure
    0.80
     reservations
    0.79
     adjustments
    0.77
     exceptions
    0.68
     arrangements
    0.68
     arrang
    0.67
    Activations Density 0.451%

    No Known Activations