© 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. 28582
    Prev
    Next
    INDEX
    Explanations

    references to the word "gle" and related variations, alongside occurrences of the term "ste" in various contexts

    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
    712
    +0.13
    0.6%
    497
    +0.11
    0.5%
    383
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.13
    0.00
    404
    +0.11
    0.02
    526
    +0.10
    0.00
    Negative Logits
    xual
    -0.76
    rians
    -0.68
    rac
    -0.65
     confir
    -0.63
     uncle
    -0.63
    DOWN
    -0.62
     acknow
    -0.60
    den
    -0.60
    ras
    -0.60
     dads
    -0.60
    POSITIVE LOGITS
    aming
    1.47
    efully
    1.33
    eful
    1.16
    amed
    1.11
    ams
    0.92
    jamin
    0.90
    atures
    0.87
    am
    0.84
    etooth
    0.84
    eless
    0.81
    Activations Density 0.085%

    No Known Activations