© 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. 29457
    Prev
    Next
    INDEX
    Explanations

    words related to associations, groups, or organizations

    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
    63
    +0.11
    0.5%
    180
    +0.09
    0.4%
    126
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.11
    -0.00
    232
    +0.09
    0.10
    659
    +0.09
    0.08
    Negative Logits
     Nebula
    -0.80
    theless
    -0.70
     Hubble
    -0.70
     stride
    -0.69
     Roose
    -0.68
    é¾įå¥ij士
    -0.68
     iceberg
    -0.66
     parchment
    -0.65
    è£ħ
    -0.65
     silence
    -0.65
    POSITIVE LOGITS
    istries
    1.02
    astery
    0.98
     Ltd
    0.93
    lege
    0.90
    uctions
    0.89
    aternity
    0.88
    eworks
    0.88
    .,
    0.87
    iffs
    0.85
    fg
    0.84
    Activations Density 1.138%

    No Known Activations