© 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. 28163
    Prev
    Next
    INDEX
    Explanations

    questions and phrases that inquire about measurements or evaluations

    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
    +0.14
    0.6%
    528
    +0.11
    0.5%
    557
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    483
    +0.14
    0.04
    662
    +0.11
    0.05
    559
    +0.10
    0.03
    Negative Logits
    GY
    -0.67
    ude
    -0.67
    tl
    -0.67
    iner
    -0.63
    ??
    -0.62
    sic
    -0.62
    uminati
    -0.62
     mistake
    -0.61
     stupidity
    -0.61
     Anyway
    -0.58
    POSITIVE LOGITS
     evenly
    1.01
     frequently
    0.99
     Flavoring
    0.90
     often
    0.90
     likely
    0.86
     many
    0.86
     much
    0.83
     actively
    0.83
     favorably
    0.83
     quickly
    0.83
    Activations Density 0.400%

    No Known Activations