© 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. 4-RES-JB
    11. 14624
    Prev
    Next
    INDEX
    Explanations

    references to physical locations or structures, particularly booths

    oai_token-act-pair · gpt-3.5-turbo

    mentions of booths

    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
    itals
    -0.71
    tymology
    -0.69
    mitting
    -0.68
    mitted
    -0.66
    inse
    -0.65
    ochond
    -0.63
     acronym
    -0.61
     acron
    -0.60
    thodox
    -0.60
    utenberg
    -0.58
    POSITIVE LOGITS
     booth
    1.15
     booths
    0.96
    holder
    0.90
    cloth
    0.89
    holders
    0.88
    washer
    0.86
    wagon
    0.80
    handler
    0.79
    swer
    0.75
     linen
    0.75
    Activations Density 0.006%

    No Known Activations