© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 8-RES_POST_32K-OAI
    11. 4664
    Prev
    Next
    INDEX
    Explanations

    references to different types of exercises targeting specific muscle groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    474
    +0.11
    0.5%
    632
    +0.10
    0.5%
    424
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.11
    0.09
    271
    +0.10
    0.05
    738
    +0.09
    0.04
    Negative Logits
     Merchants
    -0.89
    Fax
    -0.84
    Property
    -0.80
     Aad
    -0.80
    DragonMagazine
    -0.80
    agate
    -0.78
     Privacy
    -0.78
    irus
    -0.77
     Settlement
    -0.75
     indemn
    -0.75
    POSITIVE LOGITS
     cardio
    1.39
     workouts
    1.38
     exercises
    1.36
     workout
    1.34
     muscles
    1.27
     drills
    1.25
     yoga
    1.25
     squats
    1.23
     reps
    1.14
     athletes
    1.14
    Activations Density 3.123%

    No Known Activations