© 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. 27053
    Prev
    Next
    INDEX
    Explanations

    references to attitudes and their strength

    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
    247
    +0.12
    0.5%
    575
    +0.09
    0.4%
    188
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    613
    +0.12
    0.04
    759
    +0.09
    0.04
    584
    +0.09
    0.04
    Negative Logits
    enegger
    -0.80
    icles
    -0.78
    aban
    -0.73
    lan
    -0.72
    icle
    -0.70
    DERR
    -0.70
    ding
    -0.70
    istani
    -0.69
    orah
    -0.67
    earances
    -0.67
    POSITIVE LOGITS
     toward
    1.54
     towards
    1.46
     Towards
    1.18
     Tow
    1.00
     attitude
    0.94
     shift
    0.79
     attitudes
    0.77
     modification
    0.73
     change
    0.72
     reversal
    0.69
    Activations Density 0.277%

    No Known Activations