© 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. 28351
    Prev
    Next
    INDEX
    Explanations

    expressions of gratitude and support in relationships

    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
    477
    +0.12
    0.5%
    697
    +0.10
    0.5%
    482
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    354
    +0.12
    0.07
    551
    +0.10
    0.07
    566
    +0.10
    0.06
    Negative Logits
     CFR
    -0.71
     Crim
    -0.64
     Revision
    -0.63
     Criminal
    -0.62
     scram
    -0.61
    odox
    -0.61
     Arrest
    -0.61
    react
    -0.59
     unsolved
    -0.58
    entary
    -0.58
    POSITIVE LOGITS
    ©¶æ¥µ
    1.02
     afforded
    0.92
     gained
    0.90
    ãĥij
    0.90
     garnered
    0.87
     garner
    0.85
     derive
    0.82
     Gain
    0.80
     deserve
    0.80
     gain
    0.78
    Activations Density 0.888%

    No Known Activations