© 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. 6261
    Prev
    Next
    INDEX
    Explanations

    words associated with positive experiences or high energy

    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
    659
    +0.11
    0.5%
    18
    +0.11
    0.5%
    474
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.10
    284
    +0.11
    0.08
    232
    +0.11
    0.09
    Negative Logits
    Manufact
    -0.80
     Airbus
    -0.75
    Graphics
    -0.75
    Iterator
    -0.73
     bankruptcy
    -0.73
    Buyable
    -0.71
    Reloaded
    -0.69
     abstraction
    -0.69
    ICS
    -0.69
    CLASSIFIED
    -0.68
    POSITIVE LOGITS
    vati
    1.39
    yan
    1.18
    zu
    1.14
    hai
    1.08
    nu
    1.07
    iu
    1.07
    ku
    1.05
    wi
    1.05
    dan
    1.03
    uin
    1.02
    Activations Density 0.963%

    No Known Activations