© 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. 8-RES-JB
    11. 6273
    Prev
    Next
    INDEX
    Explanations

    phrases related to established or traditional practices

    oai_token-act-pair · gpt-3.5-turbo

    references to traditional behaviors or established norms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ighters
    -0.68
    merce
    -0.67
    semble
    -0.65
    UST
    -0.64
    scl
    -0.63
     Chic
    -0.62
    hma
    -0.62
    endez
    -0.61
    abella
    -0.61
    mingham
    -0.60
    POSITIVE LOGITS
     whereby
    1.26
    pmwiki
    0.93
    ually
    0.90
     tendency
    0.85
     refrain
    0.78
     reversal
    0.77
    horse
    0.75
     wherein
    0.74
     resorted
    0.69
     habit
    0.68
    Activations Density 0.183%

    No Known Activations