© 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. 11293
    Prev
    Next
    INDEX
    Explanations

    words related to the concept of 'lifting' or 'raising'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "lift" in various forms

    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
     alliances
    -0.73
     Tul
    -0.68
    rade
    -0.62
    course
    -0.61
    nel
    -0.60
    interest
    -0.60
    rex
    -0.58
    aval
    -0.57
    develop
    -0.57
     Delaware
    -0.57
    POSITIVE LOGITS
     lifted
    3.62
     lifting
    2.01
     lifts
    1.95
     lift
    1.84
     eased
    1.60
     Lift
    1.44
     lowered
    1.30
     removed
    1.19
     slapped
    1.18
     pulled
    1.18
    Activations Density 0.006%

    No Known Activations