© 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. 28125
    Prev
    Next
    INDEX
    Explanations

    references to specific locations and organizations

    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
    +0.12
    0.6%
    50
    +0.12
    0.5%
    174
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.12
    -0.02
    659
    +0.12
    0.02
    129
    +0.10
    0.01
    Negative Logits
    ascript
    -0.76
    Tokens
    -0.64
    4000
    -0.64
    dozen
    -0.64
    azon
    -0.61
    nih
    -0.61
    ulhu
    -0.60
     âī¡
    -0.58
    400
    -0.57
    200000
    -0.56
    POSITIVE LOGITS
    osa
    0.68
    ilib
    0.68
    llo
    0.67
     Manager
    0.65
     Shooter
    0.64
    oulos
    0.62
    isal
    0.61
    ilus
    0.60
    agher
    0.59
     Bagg
    0.58
    Activations Density 0.160%

    No Known Activations