© 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. 5231
    Prev
    Next
    INDEX
    Explanations

    instances of the word "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
    373
    +0.13
    0.6%
    233
    +0.10
    0.5%
    274
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    616
    +0.13
    0.02
    341
    +0.10
    0.01
    637
    +0.09
    0.02
    Negative Logits
     Reincarn
    -0.71
     preced
    -0.65
     procedure
    -0.63
     Citation
    -0.62
    gaard
    -0.61
     scrimmage
    -0.58
     exceptions
    -0.58
    cycles
    -0.57
     Procedure
    -0.57
     exception
    -0.56
    POSITIVE LOGITS
    clusive
    0.97
    soever
    0.69
    auri
    0.65
    ighter
    0.65
    ahu
    0.64
    oots
    0.63
     sudden
    0.62
    ooting
    0.62
    nuts
    0.61
    CLUS
    0.60
    Activations Density 0.162%

    No Known Activations