© 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. 5177
    Prev
    Next
    INDEX
    Explanations

    references to political and social issues, particularly regarding rights, regulations, and community dynamics

    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
    741
    +0.09
    0.4%
    155
    +0.09
    0.4%
    193
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.09
    0.35
    575
    +0.09
    0.25
    526
    +0.09
    -0.14
    Negative Logits
     AND
    -0.68
     and
    -0.67
    ibrary
    -0.59
     And
    -0.56
    And
    -0.55
    Scene
    -0.55
     Oo
    -0.50
    and
    -0.49
     Category
    -0.48
    DonaldTrump
    -0.48
    POSITIVE LOGITS
     thereof
    1.27
     accordingly
    1.27
     thereafter
    1.22
     alike
    1.11
     thereto
    1.10
     therein
    1.09
     versa
    0.99
     respectively
    0.99
     afterward
    0.91
     afterwards
    0.85
    Activations Density 21.865%

    No Known Activations