© 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. 8-RES_POST_32K-OAI
    11. 8546
    Prev
    Next
    INDEX
    Explanations

    references to video game-related terminology or characters

    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
    314
    +0.14
    0.6%
    659
    +0.11
    0.5%
    277
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.14
    0.12
    314
    +0.11
    0.01
    232
    +0.09
    0.09
    Negative Logits
     textbooks
    -0.90
     financing
    -0.90
     UNESCO
    -0.82
     fundra
    -0.82
     centers
    -0.80
     repertoire
    -0.80
     closures
    -0.80
     closure
    -0.79
     foundations
    -0.79
     citations
    -0.76
    POSITIVE LOGITS
    Magikarp
    1.34
    gamer
    1.30
    bear
    1.26
    Wolf
    1.21
    Gamer
    1.20
    Gaming
    1.19
    guy
    1.16
    Knight
    1.16
    Bear
    1.13
    Guy
    1.10
    Activations Density 1.075%

    No Known Activations