© 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. 4-RES-JB
    11. 11434
    Prev
    Next
    INDEX
    Explanations

    the term "SWAT"

    oai_token-act-pair · gpt-3.5-turbo

    references to SWAT teams or related law enforcement terminology

    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
    士
    -0.91
     Minotaur
    -0.81
    Downloadha
    -0.72
     deliber
    -0.64
    infeld
    -0.64
     Sins
    -0.64
     Dominion
    -0.63
     omission
    -0.60
     capsule
    -0.60
    Ô
    -0.60
    POSITIVE LOGITS
     swat
    1.21
    ting
    1.08
    eenth
    0.95
    estic
    0.87
    efully
    0.86
    TING
    0.85
    eful
    0.84
    eteenth
    0.84
    stakes
    0.83
    ishers
    0.80
    Activations Density 0.004%

    No Known Activations