© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 29469
    Prev
    Next
    INDEX
    Explanations

    instances of bracketed or grouped mathematical expressions and notation

    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
    160
    +0.14
    0.7%
    218
    +0.13
    0.6%
    314
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.14
    0.02
    314
    +0.13
    -0.00
    107
    +0.13
    0.01
    Negative Logits
     fid
    -0.79
     Jem
    -0.78
     Fut
    -0.75
     Ago
    -0.69
     somew
    -0.67
     Ka
    -0.67
     Kum
    -0.67
     Bett
    -0.66
     thous
    -0.66
     furthe
    -0.66
    POSITIVE LOGITS
    Ice
    1.15
    Spirit
    1.11
    Dex
    1.11
    slice
    1.09
    Hun
    1.08
    sic
    1.05
    MR
    1.02
    pain
    1.01
    Ward
    0.98
    Str
    0.98
    Activations Density 0.058%

    No Known Activations