© 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. 0-RES-JB
    11. 16521
    Prev
    Next
    INDEX
    Explanations

    phrases related to employment or tasks being carried out

    oai_token-act-pair · gpt-3.5-turbo
    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
     Royale
    -0.64
     roared
    -0.64
    zeb
    -0.64
     VIDEOS
    -0.63
    ennes
    -0.62
    uden
    -0.62
    np
    -0.62
    LET
    -0.62
     invade
    -0.61
    oras
    -0.61
    POSITIVE LOGITS
     working
    3.41
     Working
    2.46
    Working
    2.44
    working
    2.41
     collaborating
    1.76
     WORK
    1.69
     worked
    1.60
     work
    1.57
     functioning
    1.48
     cooperating
    1.41
    Activations Density 0.038%

    No Known Activations