© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 9-RES-JB
    11. 23400
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "done" emphasized through various formats such as different tenses and conjugations

    oai_token-act-pair · gpt-3.5-turbo

    the word "done" and its variations, indicating completion or finished actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    olson
    -0.75
    ulates
    -0.71
    mare
    -0.66
    ulative
    -0.66
    orum
    -0.64
    lights
    -0.64
    ÏĦ
    -0.63
    ents
    -0.63
    uckle
    -0.63
    Demand
    -0.62
    POSITIVE LOGITS
    pez
    1.27
     differently
    0.77
     administr
    0.74
     away
    0.70
    chy
    0.69
     wrong
    0.68
     justice
    0.68
    ggie
    0.68
    omething
    0.66
    zed
    0.66
    Activations Density 0.052%

    No Known Activations