© 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. 8-RES-JB
    11. 7217
    Prev
    Next
    INDEX
    Explanations

    films or TV shows that are based on existing stories or source material

    oai_token-act-pair · gpt-3.5-turbo

    references to adaptations of stories or films based on existing works

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    teness
    -0.90
    acity
    -0.88
    gencies
    -0.87
    apo
    -0.81
    fw
    -0.79
    ffect
    -0.79
    alde
    -0.78
    wcs
    -0.74
    erity
    -0.73
    dash
    -0.72
    POSITIVE LOGITS
     bestselling
    0.89
     novel
    0.79
     classic
    0.79
     screenplay
    0.77
     Akira
    0.77
     popular
    0.76
     autobiography
    0.75
     copyrighted
    0.75
     novels
    0.75
     Artemis
    0.75
    Activations Density 0.207%

    No Known Activations