© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 18648
    Prev
    Next
    INDEX
    Explanations

    file extensions related to executable files and configuration files

    oai_token-act-pair · gpt-3.5-turbo

    file extensions and configuration-related terms in programming or computer science contexts

    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
     extremes
    -0.72
     ages
    -0.70
     poles
    -0.64
     pige
    -0.64
     limb
    -0.64
     sidewalks
    -0.62
    sticks
    -0.62
     rooft
    -0.61
     highs
    -0.61
     boarded
    -0.60
    POSITIVE LOGITS
    çīĪ
    0.80
    ignore
    0.79
    Directory
    0.78
    fw
    0.78
    rc
    0.76
    Daddy
    0.75
    orgetown
    0.73
     Mehran
    0.72
    olkien
    0.72
    itchie
    0.71
    Activations Density 0.030%

    No Known Activations