© 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 EvalsExportsinterp-engineNEWAPI 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. 27190
    Prev
    Next
    INDEX
    Explanations

    characters and symbols, particularly those related to programming or markup languages

    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
    665
    +0.12
    0.6%
    690
    +0.11
    0.5%
    567
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.12
    0.01
    117
    +0.11
    0.01
    613
    +0.09
    0.01
    Negative Logits
    phia
    -0.81
    enance
    -0.77
     seiz
    -0.70
     opin
    -0.67
     skyline
    -0.67
    orah
    -0.66
    amines
    -0.66
    aminer
    -0.65
     contrace
    -0.64
     occas
    -0.63
    POSITIVE LOGITS
    ä¼
    1.05
    ä
    0.99
    å¥
    0.83
    ĨĴ
    0.80
    å®
    0.76
    yssey
    0.76
    ł
    0.75
    LOCK
    0.75
     Puzzles
    0.74
    è¿
    0.74
    Activations Density 0.035%

    No Known Activations