© 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. 19149
    Prev
    Next
    INDEX
    Explanations

    terms related to robots and machines

    oai_token-act-pair · gpt-3.5-turbo

    repeated instances of a specific term referring to automated entities or machines

    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
     Wiley
    -0.69
     competitor
    -0.68
     MD
    -0.64
     fitness
    -0.61
     chast
    -0.61
    backer
    -0.61
     predecessor
    -0.60
     client
    -0.59
     collision
    -0.58
     abuse
    -0.58
    POSITIVE LOGITS
    ots
    4.62
    OTS
    2.39
    ot
    2.19
    otted
    2.02
    oted
    1.73
    oting
    1.72
    otes
    1.64
    OT
    1.61
    otics
    1.33
    ott
    1.32
    Activations Density 0.008%

    No Known Activations