© 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. 5142
    Prev
    Next
    INDEX
    Explanations

    specific formatting or structure in text, such as punctuation marks and symbols used in data representation

    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
    379
    +0.10
    0.5%
    372
    +0.10
    0.4%
    499
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.10
    -0.03
    555
    +0.10
    0.09
    107
    +0.09
    0.09
    Negative Logits
    uta
    -0.80
    ãĥĺ
    -0.80
     Ut
    -0.72
     Unity
    -0.72
    IF
    -0.71
     Tradable
    -0.70
     Deliver
    -0.69
     uncond
    -0.68
     Seah
    -0.68
     expend
    -0.68
    POSITIVE LOGITS
    zer
    1.03
    zos
    0.99
    zing
    0.97
    zed
    0.96
    zo
    0.95
    jer
    0.93
    z
    0.89
    zik
    0.88
    zers
    0.86
    morph
    0.86
    Activations Density 4.210%

    No Known Activations