© 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. 28821
    Prev
    Next
    INDEX
    Explanations

    phrases indicating questions or comparisons related to behavior and perception

    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
    497
    +0.12
    0.5%
    687
    +0.11
    0.5%
    35
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.12
    0.14
    285
    +0.11
    0.07
    297
    +0.10
    0.08
    Negative Logits
    orsche
    -0.69
    FIELD
    -0.68
    ahime
    -0.66
    ãĤº
    -0.64
    ãĥŁ
    -0.63
    UGE
    -0.60
    nces
    -0.57
    KEN
    -0.57
    oute
    -0.57
     inclusion
    -0.57
    POSITIVE LOGITS
     how
    1.50
    how
    1.39
     How
    1.31
     HOW
    1.16
    How
    1.13
    why
    1.04
     why
    1.04
    what
    1.03
     what
    1.00
    HOW
    0.97
    Activations Density 1.908%

    No Known Activations