© 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. 6-RES-JB
    11. 9340
    Prev
    Next
    INDEX
    Explanations

    phrases related to disciplinary actions or conflicts

    oai_token-act-pair · gpt-3.5-turbo

    actions involving failure to comply with rules or regulations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    endif
    -0.77
    ciating
    -0.73
    depending
    -0.71
    inger
    -0.68
    fortunately
    -0.68
    gradation
    -0.67
    erm
    -0.66
    cking
    -0.66
    assuming
    -0.65
    atari
    -0.65
    POSITIVE LOGITS
     sake
    0.87
     illegal
    0.79
     improper
    0.76
     purposes
    0.73
     nonviolent
    0.64
     unlawful
    0.63
     insensitive
    0.63
    tein
    0.62
     withd
    0.62
     unsafe
    0.62
    Activations Density 0.162%

    No Known Activations