© 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. 8-RES_POST_32K-OAI
    11. 7382
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "ky" 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
    313
    +0.13
    0.6%
    699
    +0.12
    0.5%
    497
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.13
    0.01
    699
    +0.12
    0.01
    563
    +0.09
    0.01
    Negative Logits
    ãĥ¼ãĥ³
    -0.69
    ========
    -0.67
    ient
    -0.66
    ENDED
    -0.66
    IBLE
    -0.65
    NRS
    -0.63
    uve
    -0.63
     Ic
    -0.61
    á½
    -0.59
     Hick
    -0.58
    POSITIVE LOGITS
    loo
    0.90
    yk
    0.87
    £ı
    0.83
    bors
    0.81
    eah
    0.78
    doms
    0.76
    imura
    0.76
    orea
    0.75
    lish
    0.75
    dom
    0.74
    Activations Density 0.057%

    No Known Activations