© 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. 2374
    Prev
    Next
    INDEX
    Explanations

    the concept of permanence or something lasting without change

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "permanent" and related concepts

    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
    edin
    -0.93
    ean
    -0.89
    gerald
    -0.83
    ymes
    -0.79
     GOODMAN
    -0.79
    rahim
    -0.77
    aucuses
    -0.77
    ramid
    -0.76
    DragonMagazine
    -0.76
    heit
    -0.73
    POSITIVE LOGITS
     residence
    0.92
     fixture
    0.90
    ITY
    0.87
     residency
    0.84
     injunction
    0.81
    manent
    0.79
     marker
    0.78
     resident
    0.78
     permanent
    0.78
     tenancy
    0.77
    Activations Density 0.026%

    No Known Activations