© 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. 6614
    Prev
    Next
    INDEX
    Explanations

    terms related to non-causality or non-linear relationships

    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
    102
    +0.15
    0.7%
    9
    +0.10
    0.5%
    373
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.15
    0.08
    616
    +0.10
    0.06
    508
    +0.10
    0.05
    Negative Logits
     Seasons
    -0.77
     sheets
    -0.73
     deeds
    -0.72
     Moments
    -0.70
     vows
    -0.69
     licences
    -0.68
     decency
    -0.67
     licenses
    -0.67
     reminder
    -0.67
     beetles
    -0.66
    POSITIVE LOGITS
    linear
    1.22
    functional
    1.09
    competitive
    1.01
    technical
    0.96
    traditional
    0.93
    violent
    0.93
    partisan
    0.92
    productive
    0.91
    capitalist
    0.90
    threatening
    0.90
    Activations Density 0.127%

    No Known Activations