© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 4731
    Prev
    Next
    INDEX
    Explanations

    references to political events and figures

    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
    154
    +0.10
    0.5%
    631
    +0.10
    0.5%
    598
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.10
    -0.02
    154
    +0.10
    0.06
    675
    +0.09
    0.05
    Negative Logits
    emale
    -0.67
     Subtle
    -0.64
     Stard
    -0.62
     Brass
    -0.62
    Silver
    -0.62
     Rounds
    -0.61
     Siber
    -0.61
     Polk
    -0.61
     Unic
    -0.60
     Brigham
    -0.60
    POSITIVE LOGITS
    independence
    0.83
     AFP
    0.79
     RTX
    0.71
     strawberries
    0.70
    ãģ®ç
    0.70
    æ©
    0.69
     Pict
    0.69
    DragonMagazine
    0.68
     celebrations
    0.68
     Dalai
    0.67
    Activations Density 1.309%

    No Known Activations