© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 6982
    Prev
    Next
    INDEX
    Explanations

    references to online video platforms and editing functions

    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
    257
    +0.12
    0.5%
    314
    +0.12
    0.5%
    81
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.12
    -0.01
    76
    +0.12
    0.06
    208
    +0.10
    0.06
    Negative Logits
    gaard
    -0.68
     Sanford
    -0.68
    arnaev
    -0.65
    ongyang
    -0.62
     Murdoch
    -0.58
     Allaah
    -0.58
     Juda
    -0.58
     Anchorage
    -0.56
     Mayer
    -0.56
    fts
    -0.56
    POSITIVE LOGITS
     ,"
    1.12
    Í
    1.04
     ",
    0.99
     ."
    0.94
     ):
    0.92
     ],
    0.91
     ),
    0.89
     )]
    0.89
     ].
    0.86
     );
    0.84
    Activations Density 2.053%

    No Known Activations