© 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. 4-RES-JB
    11. 10383
    Prev
    Next
    INDEX
    Explanations

    phrases related to fictional elements, such as fictional worlds, characters, and settings

    oai_token-act-pair · gpt-3.5-turbo

    references to fictional elements and settings

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

    No Comments

    Negative Logits
    ktop
    -0.87
    hammad
    -0.82
    gans
    -0.76
    feeding
    -0.74
    hens
    -0.72
    VL
    -0.71
    erto
    -0.70
    sterdam
    -0.70
    cler
    -0.70
    chains
    -0.69
    POSITIVE LOGITS
    ized
    1.03
    istically
    1.03
     fictional
    0.98
     portray
    0.96
     universes
    0.95
    acters
    0.94
     recre
    0.90
     portrayal
    0.89
    ization
    0.86
    izations
    0.86
    Activations Density 0.018%

    No Known Activations