© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 2-RES-JB
    11. 8053
    Prev
    Next
    INDEX
    Explanations

    words with the root "se" followed by a single letter that typically denotes strong action or emphasis

    oai_token-act-pair · gpt-3.5-turbo

    the presence of the prefix "se-" in words

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    initely
    -0.73
    raviolet
    -0.73
    enhagen
    -0.72
    INGTON
    -0.70
    enegger
    -0.69
    heid
    -0.68
    SHIP
    -0.67
    hetti
    -0.66
     hoops
    -0.66
    kson
    -0.64
    POSITIVE LOGITS
    eps
    1.03
    perate
    0.94
    leanor
    0.93
    vel
    0.89
    wed
    0.87
    aring
    0.86
    vers
    0.86
    ult
    0.85
    rend
    0.85
    eker
    0.85
    Activations Density 0.011%

    No Known Activations