© 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 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. 4-RES-JB
    11. 4798
    Prev
    Next
    INDEX
    Explanations

    words related to being physically uncovered or exposed

    oai_token-act-pair · gpt-3.5-turbo

    words and phrases associated with the concept of being "under" or "undermined."

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

    No Comments

    Negative Logits
     Flavoring
    -0.94
    anwhile
    -0.83
    azines
    -0.83
    ramid
    -0.82
    zzo
    -0.81
    ebus
    -0.81
    pheus
    -0.79
    ppo
    -0.78
     Blaze
    -0.78
    gdala
    -0.78
    POSITIVE LOGITS
    aunted
    0.90
    oubt
    0.86
    iscovered
    0.85
    ated
    0.80
    ried
    0.80
    etermined
    0.79
    ications
    0.79
    oubted
    0.79
    oub
    0.79
    wed
    0.78
    Activations Density 0.007%

    No Known Activations