© 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. 8942
    Prev
    Next
    INDEX
    Explanations

    terms associated with inclusion and encroachment

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.11
    0.5%
    223
    +0.11
    0.5%
    746
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    746
    +0.11
    0.01
    251
    +0.11
    0.01
    676
    +0.11
    0.01
    Negative Logits
     concurrent
    -0.68
     Prosper
    -0.61
     breeze
    -0.59
     Keeper
    -0.58
    zai
    -0.58
    Sil
    -0.58
     lyr
    -0.56
    efully
    -0.56
    BALL
    -0.55
     custod
    -0.55
    POSITIVE LOGITS
    outer
    1.03
    roach
    1.01
    ³
    0.97
    µ
    0.89
    į
    0.87
    rust
    0.87
    ased
    0.87
    rowing
    0.86
    ²
    0.86
    ¨
    0.84
    Activations Density 0.031%

    No Known Activations