© 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. 31636
    Prev
    Next
    INDEX
    Explanations

    recurring phrases and structures in dialogue and formal language

    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
    401
    +0.12
    0.5%
    767
    +0.09
    0.4%
    243
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    401
    +0.12
    0.06
    607
    +0.09
    0.03
    126
    +0.09
    0.03
    Negative Logits
     Glad
    -0.67
    clus
    -0.66
    sure
    -0.62
     Mai
    -0.60
    ika
    -0.60
    lime
    -0.60
    idian
    -0.59
    eve
    -0.57
    milo
    -0.56
    actionDate
    -0.56
    POSITIVE LOGITS
     Over
    1.97
     OVER
    1.81
     Under
    1.81
    over
    1.80
     over
    1.80
     under
    1.76
    Over
    1.71
    under
    1.64
     UNDER
    1.63
    Under
    1.59
    Activations Density 1.884%

    No Known Activations