© 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. 11-RES-JB
    11. 3362
    Prev
    Next
    INDEX
    Explanations

    phrases related to loss, removal, or deprivation

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "stripped" and its variations in the context of loss or removal

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    NOR
    -0.73
    enegger
    -0.71
    risome
    -0.70
    foreseen
    -0.66
    anwhile
    -0.62
    abad
    -0.62
    rosso
    -0.60
     Incident
    -0.60
    Interstitial
    -0.59
    ondon
    -0.59
    POSITIVE LOGITS
     away
    0.99
     of
    0.98
     thereof
    0.87
     naked
    0.83
     bare
    0.79
     alive
    0.74
    ysis
    0.72
     clean
    0.71
     redundant
    0.70
    ynthesis
    0.69
    Activations Density 0.084%

    No Known Activations