© 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. 6187
    Prev
    Next
    INDEX
    Explanations

    references to vintage and obsolete items or products

    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
    497
    +0.11
    0.5%
    161
    +0.10
    0.4%
    598
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.11
    0.08
    288
    +0.10
    0.08
    91
    +0.09
    0.06
    Negative Logits
    ythm
    -0.84
    jriwal
    -0.82
    agonist
    -0.81
    xual
    -0.79
    SPONSORED
    -0.78
    aminer
    -0.77
    Goal
    -0.74
    ipel
    -0.74
    ularity
    -0.73
    lement
    -0.73
    POSITIVE LOGITS
     antique
    1.08
     vintage
    0.92
     aluminum
    0.91
     typew
    0.89
     automobiles
    0.88
     Soviet
    0.85
     Victorian
    0.85
     vinyl
    0.84
     furniture
    0.84
    bilt
    0.83
    Activations Density 0.828%

    No Known Activations