© 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. 5097
    Prev
    Next
    INDEX
    Explanations

    references to names or entities

    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
    516
    +0.12
    0.5%
    280
    +0.10
    0.5%
    296
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    609
    +0.12
    0.03
    666
    +0.10
    0.02
    314
    +0.10
    -0.01
    Negative Logits
    merce
    -0.66
    ilitating
    -0.66
    ibrary
    -0.65
    çīĪ
    -0.64
    roman
    -0.63
    enegger
    -0.61
    natureconservancy
    -0.60
    estate
    -0.60
    \/\/
    -0.59
    ilitation
    -0.59
    POSITIVE LOGITS
    Shift
    0.71
    ister
    0.68
    ly
    0.68
    leness
    0.64
    lev
    0.64
    essing
    0.63
    ochond
    0.61
    anni
    0.61
    arent
    0.60
     Amount
    0.59
    Activations Density 0.249%

    No Known Activations