© 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. 1-RES-JB
    11. 20980
    Prev
    Next
    INDEX
    Explanations

    words or word endings related to actions or processes, particularly ending in "king."

    oai_token-act-pair · gpt-3.5-turbo

    forms of the word "king."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ric
    -0.66
     flown
    -0.65
     developed
    -0.64
    ß
    -0.63
    querque
    -0.63
     redistributed
    -0.62
     discharge
    -0.61
     radiation
    -0.60
    Þ
    -0.59
    ATIONAL
    -0.57
    POSITIVE LOGITS
    king
    1.46
    doms
    1.03
    ked
    0.88
    leys
    0.87
    lord
    0.87
    leon
    0.84
    patrick
    0.84
    artist
    0.84
    lift
    0.84
    lass
    0.83
    Activations Density 0.005%

    No Known Activations