© 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. 6624
    Prev
    Next
    INDEX
    Explanations

    references to generic data type definitions in programming contexts

    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
    156
    +0.11
    0.5%
    182
    +0.11
    0.5%
    226
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    426
    +0.11
    0.15
    759
    +0.11
    0.15
    352
    +0.09
    0.17
    Negative Logits
    »Ĵ
    -0.64
    aneers
    -0.61
    ATURES
    -0.60
    OHN
    -0.58
     CLR
    -0.57
    Naz
    -0.57
    giene
    -0.57
    Interested
    -0.57
    imming
    -0.55
    ãĤ¹ãĥĪ
    -0.55
    POSITIVE LOGITS
    worldly
    0.91
    catentry
    0.72
    ark
    0.60
    Setup
    0.60
    hai
    0.59
    xious
    0.58
    mega
    0.57
    icans
    0.56
    parent
    0.56
    Kids
    0.56
    Activations Density 0.882%

    No Known Activations