© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 8728
    Prev
    Next
    INDEX
    Explanations

    time-related information and timestamps

    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
    373
    +0.12
    0.6%
    273
    +0.11
    0.5%
    606
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    176
    +0.12
    0.04
    145
    +0.11
    0.04
    523
    +0.09
    0.04
    Negative Logits
     periodic
    -0.69
     likewise
    -0.64
     downstream
    -0.64
     upd
    -0.63
     lodged
    -0.62
     securely
    -0.62
     removable
    -0.61
     spons
    -0.60
     gad
    -0.60
     contiguous
    -0.60
    POSITIVE LOGITS
    ĸļ
    0.86
    mix
    0.72
    aniel
    0.72
    anza
    0.70
     Dialogue
    0.70
     <@
    0.70
    iak
    0.69
    sound
    0.67
     <+
    0.66
     Objective
    0.66
    Activations Density 0.418%

    No Known Activations