© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 8871
    Prev
    Next
    INDEX
    Explanations

    punctuation marks and symbols within technical contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    300
    +0.10
    0.5%
    181
    +0.10
    0.5%
    555
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    345
    +0.10
    0.04
    176
    +0.10
    0.02
    326
    +0.09
    0.04
    Negative Logits
    MSN
    -0.85
    Ö¼
    -0.76
     Lovecraft
    -0.74
    ãĥĩ
    -0.73
    -(
    -0.72
    ür
    -0.71
     Terran
    -0.68
     Methodist
    -0.68
    ikan
    -0.68
     Presbyterian
    -0.67
    POSITIVE LOGITS
     })
    0.84
    priority
    0.72
     });
    0.72
     constructor
    0.72
     payload
    0.71
     receivers
    0.68
     dependencies
    0.68
    overe
    0.67
    properties
    0.67
     interpreter
    0.66
    Activations Density 0.214%

    No Known Activations