© 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. 4-RES-JB
    11. 1639
    Prev
    Next
    INDEX
    Explanations

    superlatives related to quality, especially describing things as the finest

    oai_token-act-pair · gpt-3.5-turbo

    adjectives and phrases describing superiority or excellence

    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
    space
    -0.75
    itals
    -0.73
    vim
    -0.72
    blank
    -0.70
    skirts
    -0.67
    uberty
    -0.67
    arter
    -0.66
    asket
    -0.66
    idem
    -0.65
    verson
    -0.65
    POSITIVE LOGITS
     buds
    0.78
    hest
    0.77
     finest
    0.75
     understatement
    0.71
     quality
    0.71
    avorite
    0.70
     minds
    0.67
    ries
    0.66
     linen
    0.65
     adolesc
    0.65
    Activations Density 0.008%

    No Known Activations