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

    Neuronpedia

    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. 10-RES_POST_32K-OAI
    11. 29367
    Prev
    Next
    INDEX
    Explanations

    words associated with improvement and growth

    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
    416
    +0.11
    0.5%
    59
    +0.10
    0.5%
    120
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    271
    +0.11
    0.02
    253
    +0.10
    0.02
    761
    +0.10
    0.02
    Negative Logits
    cium
    -0.90
    NetMessage
    -0.75
    usha
    -0.71
    DragonMagazine
    -0.69
    itto
    -0.69
    soDeliveryDate
    -0.68
    py
    -0.68
    netflix
    -0.68
     ¥
    -0.68
    aval
    -0.67
    POSITIVE LOGITS
     upon
    0.85
     outcomes
    0.84
     relations
    0.83
    ments
    0.81
    iator
    0.79
     efficiency
    0.78
     grades
    0.75
     quality
    0.73
    entle
    0.71
     Quality
    0.70
    Activations Density 0.156%

    No Known Activations