© 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. 31118
    Prev
    Next
    INDEX
    Explanations

    instances of requests or prompts for action, such as leaving comments or reviews

    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
    52
    +0.11
    0.5%
    60
    +0.11
    0.5%
    299
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.11
    0.09
    250
    +0.11
    0.06
    328
    +0.10
    0.05
    Negative Logits
     psychiat
    -0.78
     dich
    -0.71
     unavoid
    -0.70
     continuum
    -0.68
     territ
    -0.65
     frivol
    -0.65
     inflamm
    -0.63
     intangible
    -0.63
     coron
    -0.63
     technologies
    -0.63
    POSITIVE LOGITS
     Quote
    0.91
     Favor
    0.86
     Thanks
    0.83
    tsy
    0.81
     Thank
    0.80
     GitHub
    0.78
     Github
    0.77
    irmation
    0.76
    Newsletter
    0.76
     Friend
    0.75
    Activations Density 0.621%

    No Known Activations