© 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. 20517
    Prev
    Next
    INDEX
    Explanations

    instances where someone sits down for an interview or conversation

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "sat down"

    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
    orius
    -0.70
    ¶ħ
    -0.69
    ://
    -0.64
    digy
    -0.64
    Us
    -0.62
    arily
    -0.62
    {"
    -0.62
    ught
    -0.61
    minecraft
    -0.59
    SourceFile
    -0.59
    POSITIVE LOGITS
    stairs
    1.11
     stairs
    0.75
    inar
    0.74
    hill
    0.72
    LOAD
    0.66
    hall
    0.65
     paddle
    0.64
    itate
    0.64
    cloth
    0.63
    Ü
    0.63
    Activations Density 0.012%

    No Known Activations