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

    Neuronpedia

    Jacobian LensNEW
    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. 3-RES-JB
    11. 131
    Prev
    Next
    INDEX
    Explanations

    words related to the ability to be played or used

    oai_token-act-pair · gpt-3.5-turbo

    terms related to playability and usability in gaming contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     compuls
    -0.81
    romy
    -0.81
    thodox
    -0.77
    olog
    -0.75
    arius
    -0.75
    wig
    -0.72
    othe
    -0.72
    omen
    -0.71
    omers
    -0.71
    udic
    -0.70
    POSITIVE LOGITS
     playable
    1.11
    halla
    0.96
     Demo
    0.79
    Characters
    0.72
     demo
    0.71
     usable
    0.71
    ãĥ´
    0.68
     demos
    0.67
    lihood
    0.66
    ãĤ
    0.66
    Activations Density 0.020%

    No Known Activations