© 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. 28031
    Prev
    Next
    INDEX
    Explanations

    phrases related to authority and evaluation in a professional context

    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
    256
    +0.11
    0.5%
    568
    +0.10
    0.5%
    485
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    139
    +0.11
    0.20
    762
    +0.10
    0.19
    718
    +0.10
    0.17
    Negative Logits
    Mega
    -0.75
    jug
    -0.72
     Brewing
    -0.72
     Rebellion
    -0.71
    uctions
    -0.71
    Grab
    -0.68
    Torrent
    -0.68
     grocer
    -0.68
    amaz
    -0.67
    umblr
    -0.67
    POSITIVE LOGITS
     advise
    1.19
     superv
    1.19
     testified
    1.17
     oversee
    1.13
     diagnose
    1.13
     certify
    1.12
     oversaw
    1.11
     briefed
    1.10
     inspect
    1.09
     testifying
    1.06
    Activations Density 3.224%

    No Known Activations