© 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. 8-RES_POST_32K-OAI
    11. 6172
    Prev
    Next
    INDEX
    Explanations

    phrases indicating foundational beliefs or assertions based on certain principles or ideologies

    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
    631
    +0.10
    0.4%
    182
    +0.10
    0.4%
    641
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    541
    +0.10
    0.05
    569
    +0.10
    0.04
    338
    +0.09
    0.04
    Negative Logits
     Bought
    -0.69
     Braun
    -0.66
     Burst
    -0.64
    ãĥīãĥ©
    -0.64
    married
    -0.63
     Scrib
    -0.63
     jugg
    -0.63
    ester
    -0.63
    apes
    -0.62
    ESE
    -0.61
    POSITIVE LOGITS
     solely
    1.01
     purely
    0.96
    ument
    0.89
     reliance
    0.84
     primarily
    0.82
     strictly
    0.81
     entirely
    0.81
     premise
    0.80
     firmly
    0.80
     squarely
    0.79
    Activations Density 0.198%

    No Known Activations