© 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. 8-RES_POST_32K-OAI
    11. 4132
    Prev
    Next
    INDEX
    Explanations

    numerical values or metrics

    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
    373
    +0.19
    0.9%
    237
    +0.12
    0.6%
    460
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    311
    +0.19
    0.01
    668
    +0.12
    0.02
    381
    +0.11
    0.03
    Negative Logits
    yip
    -0.88
    ledged
    -0.72
    pec
    -0.71
    obyl
    -0.68
    henko
    -0.67
    sing
    -0.67
    awei
    -0.65
    orno
    -0.65
    essen
    -0.64
    rint
    -0.64
    POSITIVE LOGITS
    ä¹
    0.71
    ":"/
    0.63
    使
    0.62
     ISI
    0.57
    steamapps
    0.57
     definitely
    0.56
     influences
    0.55
    è£ħ
    0.55
     effected
    0.54
    TEXTURE
    0.54
    Activations Density 0.131%

    No Known Activations