© 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. 10-RES_POST_32K-OAI
    11. 26407
    Prev
    Next
    INDEX
    Explanations

    mentions of various practical tasks and their related contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Neuron Alignment
    Index
    Value
    % of L₁
    373
    +0.15
    0.7%
    482
    +0.09
    0.4%
    6
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    482
    +0.15
    0.04
    176
    +0.09
    0.03
    689
    +0.09
    0.03
    Negative Logits
     respectively
    -0.86
     manifold
    -0.79
     toget
    -0.73
     effected
    -0.72
    theless
    -0.70
    erville
    -0.69
    assembly
    -0.69
    catentry
    -0.67
     manif
    -0.67
    orney
    -0.66
    POSITIVE LOGITS
     Ways
    0.97
     SHARES
    0.88
     Beware
    0.86
     Avoid
    0.84
     Know
    0.83
     Explicit
    0.81
    âĺħ
    0.81
     Reasons
    0.77
     Cheap
    0.77
     Types
    0.76
    Activations Density 0.301%

    No Known Activations