© 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. 26686
    Prev
    Next
    INDEX
    Explanations

    phrases related to providing updates and additional information

    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
    373
    +0.14
    0.7%
    269
    +0.12
    0.5%
    642
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    687
    +0.14
    0.05
    364
    +0.12
    0.04
    269
    +0.10
    0.03
    Negative Logits
    etheless
    -0.75
     ç¥ŀ
    -0.74
    lihood
    -0.71
     nomine
    -0.65
     euth
    -0.65
     discrep
    -0.64
    paren
    -0.61
    00200000
    -0.58
    efer
    -0.58
     mitigating
    -0.58
    POSITIVE LOGITS
     read
    1.02
     visit
    1.02
     check
    1.00
     subscribe
    0.94
     browse
    0.90
     watch
    0.88
    read
    0.88
     checkout
    0.87
     Head
    0.86
     scroll
    0.84
    Activations Density 0.607%

    No Known Activations