© 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 EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 11286
    Prev
    Next
    INDEX
    Explanations

    times of the day written in the format of "hour p.m."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific times of day, particularly in the format of 'p.m.'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ©¶æ
    -0.90
     metic
    -0.81
     pione
    -0.76
     warr
    -0.70
    Þ
    -0.69
     convol
    -0.65
     oun
    -0.64
     practition
    -0.64
    ufact
    -0.63
    ò
    -0.63
    POSITIVE LOGITS
    .,
    1.02
    ./
    0.90
    .;
    0.88
    .,"
    0.84
    .?
    0.82
    .
    0.82
    .:
    0.80
    eter
    0.79
    .),
    0.78
    ixie
    0.78
    Activations Density 0.013%

    No Known Activations