Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 8960
    Prev
    Next
    INDEX
    Explanations

    words related to physical exercise and fitness

    oai_token-act-pair · gpt-3.5-turbo

    references to physical fitness and workout activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cens
    -0.75
     Gree
    -0.72
    cens
    -0.72
     ann
    -0.71
    responsible
    -0.67
    amus
    -0.66
     difficulties
    -0.66
     cir
    -0.65
    IAS
    -0.65
    wikipedia
    -0.65
    POSITIVE LOGITS
     workout
    1.95
     vibe
    1.81
     standout
    1.74
     audition
    1.47
    Fit
    1.42
     workouts
    1.41
     breakout
    1.34
     Fit
    1.28
    inspired
    1.25
     WiFi
    1.15
    Activations Density 0.027%

    No Known Activations