© 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. 4-RES-JB
    11. 14428
    Prev
    Next
    INDEX
    Explanations

    words containing the suffix "-est" with a high intensity

    oai_token-act-pair · gpt-3.5-turbo

    forms of the word "est", indicating superlatives or comparisons

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

    No Comments

    Negative Logits
     tremend
    -0.82
    perty
    -0.79
    chwitz
    -0.77
    hammer
    -0.74
    veyard
    -0.73
    achev
    -0.72
    iosyn
    -0.65
    keyes
    -0.65
     epid
    -0.65
    £ı
    -0.64
    POSITIVE LOGITS
    ream
    0.94
    ruct
    0.91
    imating
    0.90
    imate
    0.90
    imated
    0.87
    imates
    0.86
    owe
    0.85
    alker
    0.85
    anding
    0.77
    reet
    0.76
    Activations Density 0.017%

    No Known Activations