© 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. 4813
    Prev
    Next
    INDEX
    Explanations

    variations of the word "ate."

    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
    59
    +0.14
    0.6%
    366
    +0.10
    0.5%
    335
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.14
    0.04
    99
    +0.10
    0.02
    711
    +0.10
    0.01
    Negative Logits
    sung
    -0.83
    enegger
    -0.81
    ingen
    -0.80
    dale
    -0.79
    ness
    -0.78
    nesses
    -0.76
    ged
    -0.73
    igans
    -0.71
    aine
    -0.70
    shire
    -0.69
    POSITIVE LOGITS
    chnology
    1.09
    llular
    0.88
    rers
    0.87
    anu
    0.78
     nucleus
    0.76
    lli
    0.70
    lier
    0.70
    vich
    0.69
    xit
    0.67
    lease
    0.66
    Activations Density 0.381%

    No Known Activations