© 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. 0-RES-JB
    11. 12911
    Prev
    Next
    INDEX
    Explanations

    verbs ending in 'icked'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "icked" and its variations, indicating a focus on themes of mischief or unpleasantness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     artic
    -0.69
     focal
    -0.65
     limits
    -0.62
     division
    -0.61
     divide
    -0.61
     redu
    -0.60
    ARB
    -0.60
     unification
    -0.60
     equival
    -0.60
     official
    -0.60
    POSITIVE LOGITS
    icked
    4.29
    icking
    2.83
    icks
    2.16
    ICK
    2.02
    icker
    1.76
    icky
    1.66
    ickers
    1.57
    ick
    1.49
    itched
    1.22
    ics
    1.21
    Activations Density 0.008%

    No Known Activations