© 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. 0-RES-JB
    11. 11784
    Prev
    Next
    INDEX
    Explanations

    phrases related to using swipe gestures

    oai_token-act-pair · gpt-3.5-turbo

    words related to navigation and interactions in digital or physical environments

    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
    mberg
    -0.77
    âĸĪâĸĪâĸĪâĸĪ
    -0.77
    ø
    -0.76
     Beau
    -0.75
    ovich
    -0.72
    viron
    -0.71
    oun
    -0.71
    duc
    -0.70
     Kel
    -0.69
    Hel
    -0.69
    POSITIVE LOGITS
     swipe
    3.82
     jab
    1.23
     PIN
    1.21
    tap
    1.20
     touchscreen
    1.16
     Tinder
    1.15
    Tap
    1.11
     flick
    1.08
     Lyft
    1.03
    PIN
    1.01
    Activations Density 0.043%

    No Known Activations