© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 20767
    Prev
    Next
    INDEX
    Explanations

    last names with the pattern "vers"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the substring "vers" in various contexts

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

    No Comments

    Negative Logits
    RAW
    -0.70
    CHR
    -0.69
    redo
    -0.69
     judges
    -0.69
    Fra
    -0.65
    ZE
    -0.64
     Pearce
    -0.64
    ľ
    -0.63
    AFTA
    -0.62
    ĻĤ
    -0.60
    POSITIVE LOGITS
    atile
    1.23
    atility
    1.17
    erver
    1.11
    hip
    1.07
    chool
    1.01
    terday
    0.97
    pect
    0.94
    sein
    0.94
    ystem
    0.93
    peed
    0.93
    Activations Density 0.011%

    No Known Activations