© 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. 27482
    Prev
    Next
    INDEX
    Explanations

    references to television and related media formats

    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
    83
    +0.13
    0.6%
    725
    +0.12
    0.5%
    302
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.07
    725
    +0.12
    0.05
    493
    +0.10
    0.04
    Negative Logits
    netflix
    -0.90
     Strongh
    -0.79
    ecast
    -0.76
     Aware
    -0.72
     Sphere
    -0.69
    Root
    -0.68
     Able
    -0.68
     Tec
    -0.67
    ebook
    -0.66
    ellen
    -0.66
    POSITIVE LOGITS
     appearances
    1.14
     personality
    1.11
     personalities
    0.99
     persona
    0.98
     interviews
    0.93
     appearance
    0.90
     advertisements
    0.89
     spots
    0.87
     publicity
    0.85
     show
    0.85
    Activations Density 0.632%

    No Known Activations