© 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. 22029
    Prev
    Next
    INDEX
    Explanations

    instances of the word "with" followed by a person or entity being interacted with or discussed

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "with."

    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
    ©¶æ¥µ
    -1.01
    fighter
    -0.68
    uploads
    -0.67
    orius
    -0.65
    ischer
    -0.63
    chin
    -0.63
    price
    -0.62
    waters
    -0.62
    ynski
    -0.61
    killed
    -0.61
    POSITIVE LOGITS
     regards
    1.17
     regard
    1.14
     respect
    0.89
    stood
    0.85
     impunity
    0.81
    draw
    0.78
     representatives
    0.77
    =-=-=-=-
    0.74
     strangers
    0.71
     him
    0.68
    Activations Density 0.078%

    No Known Activations