© 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. 0-RES-JB
    11. 15023
    Prev
    Next
    INDEX
    Explanations

    the word "witnesses" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "witnesses."

    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
    hip
    -0.82
    hower
    -0.68
    qua
    -0.67
     (£
    -0.64
     guiActiveUnfocused
    -0.63
     chic
    -0.62
    ovation
    -0.61
     subt
    -0.61
    ment
    -0.60
     Basic
    -0.59
    POSITIVE LOGITS
     witnesses
    4.03
     witness
    2.73
     Witnesses
    2.51
    Witness
    2.44
     eyewitness
    2.31
    ewitness
    1.85
     Witness
    1.84
     testimonies
    1.68
     informants
    1.57
     testimony
    1.56
    Activations Density 0.012%

    No Known Activations