© 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. 8-RES_POST_32K-OAI
    11. 8515
    Prev
    Next
    INDEX
    Explanations

    the preposition "in."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    314
    +0.17
    0.8%
    373
    +0.16
    0.8%
    719
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    123
    +0.17
    0.04
    322
    +0.16
    0.03
    750
    +0.12
    0.03
    Negative Logits
     bene
    -0.66
     annex
    -0.59
     omission
    -0.54
     indicating
    -0.54
     ascending
    -0.53
     faithfully
    -0.53
     Dahl
    -0.53
     authent
    -0.52
     benevolent
    -0.52
     cass
    -0.52
    POSITIVE LOGITS
    touch
    0.83
    ked
    0.78
    aughs
    0.75
     touch
    0.75
     trouble
    0.72
    hold
    0.72
    Touch
    0.72
    angled
    0.72
     Touch
    0.71
    osponsors
    0.70
    Activations Density 0.094%

    No Known Activations