© 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. 30198
    Prev
    Next
    INDEX
    Explanations

    references to web resources and image links

    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
    732
    +0.11
    0.5%
    298
    +0.10
    0.5%
    680
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.11
    0.17
    39
    +0.10
    0.13
    500
    +0.09
    0.08
    Negative Logits
     overturned
    -0.77
     recess
    -0.70
    åħī
    -0.69
    arthed
    -0.68
     slain
    -0.67
    worn
    -0.66
     Bundy
    -0.65
     Ramos
    -0.65
    æѦ
    -0.65
     vacated
    -0.64
    POSITIVE LOGITS
     Patreon
    1.40
     PayPal
    1.35
     Login
    1.28
     login
    1.26
     subscription
    1.22
     subscriptions
    1.22
     Email
    1.21
     FREE
    1.20
     registration
    1.17
     download
    1.15
    Activations Density 3.434%

    No Known Activations