© 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. 10-RES-JB
    11. 1200
    Prev
    Next
    INDEX
    Explanations

    phrases related to identity or aspiration

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating positive roles or identities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     appointments
    -0.78
    books
    -0.77
    scenes
    -0.77
     Elections
    -0.75
    views
    -0.72
    Shares
    -0.69
    Events
    -0.68
     arrangements
    -0.67
    ances
    -0.67
    uden
    -0.65
    POSITIVE LOGITS
     believer
    1.08
     hypocr
    1.03
     millionaire
    0.99
     martyr
    0.98
     slave
    0.95
     spectator
    0.94
     loser
    0.94
     bigot
    0.93
     part
    0.93
     victim
    0.92
    Activations Density 0.127%

    No Known Activations