© 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. 19101
    Prev
    Next
    INDEX
    Explanations

    phrases related to clicking on links

    oai_token-act-pair · gpt-3.5-turbo

    terms related to online engagement, particularly clicks and downloads

    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
    ardo
    -0.62
     Martial
    -0.62
    aka
    -0.61
     Reserve
    -0.61
     consider
    -0.61
     substitute
    -0.61
     tackle
    -0.61
    rain
    -0.60
    iva
    -0.60
     Phys
    -0.59
    POSITIVE LOGITS
     clicks
    3.66
     clicked
    2.07
     taps
    1.49
     ticks
    1.39
     downloads
    1.35
     presses
    1.26
    licks
    1.24
     installs
    1.22
     laughs
    1.18
     bites
    1.06
    Activations Density 0.029%

    No Known Activations