© 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. 27718
    Prev
    Next
    INDEX
    Explanations

    phrases related to apocalyptic themes or scenarios

    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
    75
    +0.11
    0.5%
    646
    +0.11
    0.5%
    705
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.11
    -0.00
    232
    +0.11
    0.02
    526
    +0.10
    -0.00
    Negative Logits
    ãĤ¬
    -0.68
    IGHTS
    -0.65
     Jagu
    -0.64
    ggle
    -0.63
    cycles
    -0.58
    hang
    -0.58
     Jackets
    -0.58
     traged
    -0.58
     corrid
    -0.58
    ilib
    -0.57
    POSITIVE LOGITS
    athetic
    1.23
    reci
    1.14
    rior
    1.13
    olitical
    1.10
    ople
    1.05
    athy
    1.04
    ocry
    1.00
    ocaly
    0.93
    rehensive
    0.92
    rehend
    0.90
    Activations Density 0.161%

    No Known Activations