© 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. 26940
    Prev
    Next
    INDEX
    Explanations

    mathematical symbols and notations related to logic and analytics

    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
    512
    +0.12
    0.5%
    235
    +0.11
    0.5%
    373
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.12
    -0.00
    289
    +0.11
    0.01
    293
    +0.10
    0.02
    Negative Logits
     succession
    -0.70
     Mavericks
    -0.68
    ufact
    -0.65
    izable
    -0.65
    aturated
    -0.65
     Rapids
    -0.64
     Leopard
    -0.62
     Cherokee
    -0.62
     TAM
    -0.62
     marrow
    -0.60
    POSITIVE LOGITS
    ε
    1.24
    ο
    1.22
    α
    1.22
    ι
    1.19
    ī
    1.13
    á½
    1.11
    s
    1.09
    ķ
    1.08
    ĩ
    1.06
    Ĩ
    1.06
    Activations Density 0.290%

    No Known Activations