© 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. 10-RES_POST_32K-OAI
    11. 28599
    Prev
    Next
    INDEX
    Explanations

    words associated with bleeding and related medical conditions

    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
    571
    +0.15
    0.7%
    133
    +0.12
    0.6%
    377
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    571
    +0.15
    0.02
    371
    +0.12
    0.02
    225
    +0.10
    0.02
    Negative Logits
    wcsstore
    -0.75
    inqu
    -0.71
    iazep
    -0.70
    bnb
    -0.66
    ãģ«
    -0.65
    ioned
    -0.64
    ////
    -0.62
    entric
    -0.62
    _-
    -0.62
    afort
    -0.62
    POSITIVE LOGITS
     prof
    1.00
     wounds
    0.81
    heart
    0.81
    thirst
    0.77
     clot
    0.76
    eding
    0.73
    drop
    0.71
     hemorrh
    0.70
    emic
    0.69
     heart
    0.68
    Activations Density 0.084%

    No Known Activations