© 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. 31849
    Prev
    Next
    INDEX
    Explanations

    references to user interaction and submissions related to newsletters and data collection

    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
    373
    +0.19
    0.9%
    756
    +0.10
    0.5%
    737
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.19
    -0.02
    194
    +0.10
    0.02
    613
    +0.10
    0.02
    Negative Logits
    ournal
    -0.74
     totality
    -0.68
     tremend
    -0.67
    itton
    -0.65
    uthor
    -0.65
    ©¶æ
    -0.63
    erenn
    -0.62
     Mutual
    -0.60
    rational
    -0.60
    iaries
    -0.59
    POSITIVE LOGITS
    !--
    0.60
    appa
    0.56
    cia
    0.52
    !'"
    0.52
     iTunes
    0.52
    raft
    0.52
    Kenn
    0.51
    LES
    0.51
    mpire
    0.51
    itate
    0.51
    Activations Density 0.290%

    No Known Activations