© 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. 1-RES-JB
    11. 3785
    Prev
    Next
    INDEX
    Explanations

    instances of confusion or mistaken identity

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "confused."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inth
    -0.74
    ©¶æ
    -0.71
    bors
    -0.68
    ighth
    -0.67
    ILA
    -0.66
    events
    -0.62
    home
    -0.62
    HEAD
    -0.61
    projects
    -0.61
    igers
    -0.60
    POSITIVE LOGITS
     confused
    1.17
     confuse
    1.07
     bewild
    0.91
     confusion
    0.90
    ingly
    0.89
     baffled
    0.86
     Leilan
    0.84
    fully
    0.82
     misunderstand
    0.81
     misled
    0.81
    Activations Density 0.010%

    No Known Activations