© 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. 8-RES_POST_32K-OAI
    11. 6580
    Prev
    Next
    INDEX
    Explanations

    phrases indicating the use of the word "to" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    620
    +0.10
    0.4%
    760
    +0.10
    0.4%
    390
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.10
    0.14
    362
    +0.10
    0.13
    616
    +0.09
    0.08
    Negative Logits
    aeda
    -0.61
    addafi
    -0.58
    ENTS
    -0.57
    apolis
    -0.52
    antly
    -0.52
    arch
    -0.51
    ety
    -0.50
    OV
    -0.50
    DEN
    -0.50
    ovember
    -0.49
    POSITIVE LOGITS
     which
    2.09
    which
    1.82
     whom
    1.52
     Which
    1.50
    Which
    1.40
     whose
    1.39
    whose
    1.26
     whence
    1.19
     wherein
    1.16
     whereby
    0.97
    Activations Density 2.270%

    No Known Activations