© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 2-RES-JB
    11. 20871
    Prev
    Next
    INDEX
    Explanations

    words related to craftsmanship and manual work

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "work" in various contexts related to tasks or activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     colon
    -0.69
     subdued
    -0.62
     distant
    -0.59
     subdu
    -0.59
     cor
    -0.59
     venerable
    -0.58
     Sec
    -0.58
     constitu
    -0.58
     prose
    -0.57
     ringing
    -0.57
    POSITIVE LOGITS
    work
    1.17
    bench
    1.02
    station
    1.00
    fare
    0.97
    works
    0.96
    horse
    0.95
    WORK
    0.94
    place
    0.92
    ername
    0.92
    rentice
    0.91
    Activations Density 0.010%

    No Known Activations