© 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. 30208
    Prev
    Next
    INDEX
    Explanations

    instances of absence or negation related to individuals and events

    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
    363
    +0.12
    0.5%
    354
    +0.10
    0.5%
    415
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    762
    +0.12
    0.12
    139
    +0.10
    0.10
    354
    +0.10
    0.09
    Negative Logits
    ãĤ¡
    -0.81
     disproportion
    -0.75
     Infinite
    -0.74
    ãĥ³ãĤ¸
    -0.72
     placeholder
    -0.71
    tiny
    -0.70
     doomed
    -0.69
     inev
    -0.68
    è¦ļéĨĴ
    -0.68
     Invalid
    -0.67
    POSITIVE LOGITS
     bother
    0.88
     visits
    0.86
     visit
    0.83
    contact
    0.77
     contact
    0.76
     speak
    0.75
     contacting
    0.75
     attend
    0.73
     return
    0.73
    fax
    0.72
    Activations Density 1.483%

    No Known Activations