© 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. 10-RES_POST_32K-OAI
    11. 30789
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Now."

    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
    155
    +0.15
    0.7%
    267
    +0.12
    0.6%
    373
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    155
    +0.15
    0.05
    267
    +0.12
    0.04
    715
    +0.10
    0.04
    Negative Logits
     neighb
    -0.76
    BuyableInstoreAndOnline
    -0.64
    MpServer
    -0.64
     conduc
    -0.60
    natureconservancy
    -0.60
     vulner
    -0.59
    mid
    -0.59
     ment
    -0.58
    "},"
    -0.58
    feature
    -0.57
    POSITIVE LOGITS
    adays
    1.71
    here
    1.34
     imagine
    1.28
     suppose
    1.15
     compare
    1.03
     let
    0.98
     secondly
    0.92
     lets
    0.91
     listen
    0.91
     consider
    0.90
    Activations Density 0.075%

    No Known Activations