© 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. 4073
    Prev
    Next
    INDEX
    Explanations

    astrological terms or beliefs

    oai_token-act-pair · gpt-3.5-turbo

    concepts and terms related to astrology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
     chunks
    -0.75
     shared
    -0.72
     iP
    -0.69
     unden
    -0.65
     canvas
    -0.63
     assault
    -0.62
     leaps
    -0.62
     slices
    -0.62
     cans
    -0.61
     Hatch
    -0.61
    POSITIVE LOGITS
    rology
    4.39
    rolog
    4.23
    rol
    2.01
    roman
    1.50
    rosis
    1.22
    ritis
    1.19
    rogram
    1.13
    rog
    1.12
    rogen
    1.12
    rine
    1.07
    Activations Density 0.017%

    No Known Activations