© 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. 0-RES-JB
    6. 18269
    Prev
    Next
    INDEX
    Explanations

    words related to misspellings

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with spelling and grammar

    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
     Gy
    -0.68
    isodes
    -0.68
    tra
    -0.67
    machine
    -0.67
     Hung
    -0.66
    keye
    -0.65
    ebted
    -0.64
    interstitial
    -0.63
     Taiwanese
    -0.63
    Hung
    -0.63
    POSITIVE LOGITS
    pell
    3.34
     Pens
    1.78
     pens
    1.32
     vigil
    1.26
     QC
    1.25
    quin
    1.20
     Parish
    1.19
     Riv
    1.13
     Chronicles
    1.03
     redundant
    1.02
    Activations Density 0.060%

    No Known Activations