© 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. 1528
    Prev
    Next
    INDEX
    Explanations

    verbs related to physical actions or interactions, especially related to exerting force or influence

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "tap" and its variations in context

    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
     LDL
    -0.71
    aine
    -0.70
    itals
    -0.70
     HDL
    -0.69
    VM
    -0.63
    imon
    -0.63
    vm
    -0.63
    iveness
    -0.62
    qa
    -0.62
    oler
    -0.62
    POSITIVE LOGITS
     tapped
    3.44
     tapping
    3.27
     taps
    3.00
     tap
    2.36
    tap
    2.23
    Tap
    1.95
     Tap
    1.72
     wiret
    1.13
     siph
    1.07
     swipe
    0.98
    Activations Density 0.015%

    No Known Activations