© 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. 0-RES-JB
    11. 19252
    Prev
    Next
    INDEX
    Explanations

    words related to legal testimonies or accounts

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "witness."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    pop
    -0.72
    inyl
    -0.67
    anks
    -0.63
     (£
    -0.61
    Menu
    -0.60
     Basic
    -0.60
     Chop
    -0.59
    oku
    -0.59
     ($
    -0.59
    rats
    -0.58
    POSITIVE LOGITS
     witness
    4.16
     witnesses
    2.65
     Witness
    2.46
    Witness
    2.33
     eyewitness
    1.99
     Witnesses
    1.80
     witnessing
    1.72
     testimony
    1.68
    ewitness
    1.53
     testify
    1.49
    Activations Density 0.019%

    No Known Activations