© 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 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. 11-RES-JB
    11. 15046
    Prev
    Next
    INDEX
    Explanations

    contractions of "do not", especially with high importance on instances where the contraction "don't" is used

    oai_token-act-pair · gpt-3.5-turbo

    negations or forms of the word "don't."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     reluct
    -0.97
     newcom
    -0.95
     exha
    -0.93
     enthusi
    -0.93
    Þ
    -0.91
     pione
    -0.91
    aditional
    -0.88
     challeng
    -0.88
     princ
    -0.86
     conclud
    -0.85
    POSITIVE LOGITS
    't
    1.62
    ned
    1.18
    ning
    1.03
    ates
    0.93
    uts
    0.92
    keys
    0.84
    ate
    0.84
    ´
    0.83
    \'
    0.82
    ners
    0.81
    Activations Density 0.113%

    No Known Activations