© 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. 5358
    Prev
    Next
    INDEX
    Explanations

    directories or folder names in a computer-related context

    oai_token-act-pair · gpt-3.5-turbo

    mentions of directories and their structures

    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
     Twist
    -0.72
    oil
    -0.70
     Kah
    -0.70
     Nieto
    -0.69
     Neck
    -0.68
    fest
    -0.66
     Reaction
    -0.65
    olph
    -0.65
    neck
    -0.65
     Prot
    -0.65
    POSITIVE LOGITS
     directory
    3.54
     directories
    3.01
     Directory
    2.84
    directory
    2.67
    Directory
    2.38
     folder
    1.41
    DIR
    1.33
    dir
    1.28
     database
    1.27
     filesystem
    1.25
    Activations Density 0.020%

    No Known Activations