© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 30271
    Prev
    Next
    INDEX
    Explanations

    instances of the phrase "Someone in Red Dead Redemption" and specific references to related content and versions

    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
    322
    +0.09
    0.4%
    186
    +0.09
    0.4%
    241
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    442
    +0.09
    0.04
    107
    +0.09
    0.04
    192
    +0.09
    0.02
    Negative Logits
     rounding
    -0.83
     doubling
    -0.82
     amplification
    -0.75
     acceleration
    -0.74
     increment
    -0.74
     ringing
    -0.72
     compounded
    -0.72
     multiplying
    -0.71
     eleph
    -0.71
     Teg
    -0.70
    POSITIVE LOGITS
    DragonMagazine
    1.05
    Movie
    1.03
    Files
    1.03
    Images
    1.00
    _-_
    0.99
    ovies
    0.96
    Player
    0.90
    ideos
    0.89
    Series
    0.89
    Film
    0.88
    Activations Density 0.130%

    No Known Activations