© 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. 7413
    Prev
    Next
    INDEX
    Explanations

    references to French names or terms

    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
    613
    +0.12
    0.6%
    549
    +0.10
    0.5%
    78
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    316
    +0.12
    0.01
    13
    +0.10
    0.01
    28
    +0.10
    0.01
    Negative Logits
    avers
    -0.78
     redund
    -0.72
    onday
    -0.71
    ngth
    -0.71
    bourg
    -0.70
     ancest
    -0.66
     Protector
    -0.63
    hips
    -0.63
     Passage
    -0.63
    IFIC
    -0.62
    POSITIVE LOGITS
    ãĥ¼ãĤ¯
    0.88
    ãĥ³
    0.79
    ted
    0.78
    ãĥ«
    0.78
    ãĥį
    0.77
    untarily
    0.71
    lex
    0.71
    ãĤ¤
    0.70
    ãĥ¼ãĥ«
    0.70
    Ñĥ
    0.69
    Activations Density 0.018%

    No Known Activations