© 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. 10-RES_POST_32K-OAI
    11. 27192
    Prev
    Next
    INDEX
    Explanations

    phrases indicating help, support, or assistance

    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
    192
    +0.12
    0.5%
    392
    +0.10
    0.5%
    608
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.12
    0.12
    362
    +0.10
    0.10
    647
    +0.10
    0.06
    Negative Logits
     proble
    -0.81
    UME
    -0.74
    angles
    -0.72
     isEnabled
    -0.72
    ãĤ¨ãĥ«
    -0.72
     denies
    -0.71
    Ń·
    -0.71
    viol
    -0.70
    terness
    -0.70
    erenn
    -0.70
    POSITIVE LOGITS
     generous
    1.29
     hindsight
    1.06
     diligent
    1.00
     assistance
    0.98
     timely
    0.96
     plentiful
    0.94
     donations
    0.94
     clever
    0.94
     supportive
    0.94
     advancements
    0.93
    Activations Density 0.994%

    No Known Activations