© 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. 544
    Prev
    Next
    INDEX
    Explanations

    musical terms or entities, specifically related to the music genre or elements of "Sonata"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of a particular individual or character named "Son."

    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
    aligned
    -0.83
    UFF
    -0.74
    ilateral
    -0.68
    ille
    -0.67
    inance
    -0.61
     chests
    -0.60
    GROUND
    -0.59
    RA
    -0.58
    confidence
    -0.58
    ILLE
    -0.57
    POSITIVE LOGITS
     Son
    3.99
    Son
    3.05
     Sons
    2.08
     son
    1.73
     Daughter
    1.68
     sons
    1.36
     Father
    1.30
    son
    1.30
     Sonny
    1.20
    SON
    1.15
    Activations Density 0.013%

    No Known Activations