© 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. 10-RES-JB
    11. 18812
    Prev
    Next
    INDEX
    Explanations

    similes that express a comparison between different situations or experiences

    oai_token-act-pair · gpt-3.5-turbo

    similes or comparisons using the phrase "like."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    iencies
    -0.81
    iets
    -0.75
    erity
    -0.72
    icators
    -0.71
    ixel
    -0.70
    icipated
    -0.66
    iform
    -0.65
    IAL
    -0.65
    icals
    -0.65
    Redditor
    -0.64
    POSITIVE LOGITS
    lier
    1.32
    liest
    1.20
     watching
    0.91
     waking
    0.89
    lihood
    0.88
     comparing
    0.86
     stepping
    0.85
     discovering
    0.85
     rubbing
    0.80
     spitting
    0.80
    Activations Density 0.073%

    No Known Activations