© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 24555
    Prev
    Next
    INDEX
    Explanations

    phrases related to satisfaction or approval

    oai_token-act-pair · gpt-3.5-turbo

    expressions related to gratitude and the concept of being gratuitous

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Construct
    -0.76
    uyomi
    -0.67
     Breaker
    -0.66
     Formation
    -0.66
     Tycoon
    -0.65
     Equity
    -0.63
     Predict
    -0.62
     Universities
    -0.61
     Hemp
    -0.61
     Dwell
    -0.60
    POSITIVE LOGITS
    uitous
    1.49
     grat
    1.21
    uit
    1.03
    iously
    0.99
    uity
    0.95
    uously
    0.93
    iless
    0.90
    ified
    0.90
    ãĥ¼ãĤ¯
    0.89
    tesy
    0.84
    Activations Density 0.013%

    No Known Activations