© 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. 21369
    Prev
    Next
    INDEX
    Explanations

    computer-related terms or actions related to file management

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "file"

    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
    ĵĺ
    -0.80
     Palestin
    -0.68
    akening
    -0.68
    rontal
    -0.66
    okin
    -0.64
    ilipp
    -0.64
    orical
    -0.63
    ¿½
    -0.63
    udeb
    -0.62
    angular
    -0.62
    POSITIVE LOGITS
     FILE
    1.07
     file
    1.05
    file
    1.04
     File
    1.00
    Upload
    0.99
    ettes
    0.95
     files
    0.92
    File
    0.87
    files
    0.87
    ette
    0.86
    Activations Density 0.015%

    No Known Activations