© 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. 10-RES_POST_32K-OAI
    11. 32623
    Prev
    Next
    INDEX
    Explanations

    references to the mineral "ore" and related terms in various contexts

    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
    464
    +0.15
    0.7%
    37
    +0.10
    0.5%
    567
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    726
    +0.15
    0.01
    682
    +0.10
    0.01
    574
    +0.10
    0.00
    Negative Logits
    enegger
    -0.96
    vernment
    -0.80
     largeDownload
    -0.79
    heit
    -0.79
    -+-+
    -0.78
     Noir
    -0.77
     [/
    -0.74
    OHN
    -0.72
    ··
    -0.70
    -+
    -0.68
    POSITIVE LOGITS
     disposal
    0.61
     checkout
    0.60
     sinking
    0.59
     worn
    0.59
     guiActiveUnfocused
    0.58
    ict
    0.57
     attached
    0.57
    ics
    0.55
     invoice
    0.54
    bowl
    0.54
    Activations Density 0.080%

    No Known Activations