© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4859
    Prev
    Next
    INDEX
    Explanations

    occurrences of specific numerical and structured elements within text

    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
    314
    +0.14
    0.6%
    118
    +0.11
    0.5%
    174
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    83
    +0.14
    0.01
    178
    +0.11
    0.01
    452
    +0.10
    0.01
    Negative Logits
    urers
    -0.75
    abouts
    -0.74
    lihood
    -0.73
    aband
    -0.70
    ebook
    -0.69
    owsky
    -0.68
    glers
    -0.68
    erity
    -0.68
    devices
    -0.67
    lets
    -0.66
    POSITIVE LOGITS
    ItemImage
    0.79
     Fein
    0.61
     attRot
    0.61
    ussian
    0.60
    OGR
    0.57
     cosmos
    0.56
    perate
    0.56
     appl
    0.56
     reflected
    0.55
     Instr
    0.55
    Activations Density 0.048%

    No Known Activations