© 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. 0-RES-JB
    11. 18162
    Prev
    Next
    INDEX
    Explanations

    references to dates, schedules, and events in a calendar

    oai_token-act-pair · gpt-3.5-turbo

    references to calendars or scheduling events

    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
    orge
    -0.76
    vant
    -0.73
    unk
    -0.73
    aku
    -0.65
    unks
    -0.65
    lesh
    -0.61
    ought
    -0.61
     overpower
    -0.61
    eal
    -0.61
    rust
    -0.61
    POSITIVE LOGITS
     calendar
    3.97
     calendars
    3.00
     Calendar
    2.92
     schedule
    1.68
    endars
    1.43
     clock
    1.33
     schedules
    1.31
     timetable
    1.27
    endar
    1.26
     scheduling
    1.24
    Activations Density 0.008%

    No Known Activations