© 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. 32338
    Prev
    Next
    INDEX
    Explanations

    instances of the article "a"

    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
    724
    +0.11
    0.5%
    610
    +0.10
    0.4%
    478
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.11
    0.05
    263
    +0.10
    0.03
    319
    +0.09
    0.03
    Negative Logits
    ierrez
    -0.71
    ModLoader
    -0.58
     strangely
    -0.58
     Brav
    -0.56
     wisely
    -0.56
     mathemat
    -0.56
     conduc
    -0.54
     Progress
    -0.54
     reluct
    -0.54
    adobe
    -0.53
    POSITIVE LOGITS
    dozen
    1.53
     dozen
    1.47
     decade
    1.25
     hundred
    1.19
     thousand
    1.14
     month
    1.14
     year
    1.11
     thirds
    1.10
     century
    1.10
     hour
    1.07
    Activations Density 0.152%

    No Known Activations