© 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. 0-RES-JB
    11. 9463
    Prev
    Next
    INDEX
    Explanations

    charismatic and handsome individuals

    oai_token-act-pair · gpt-3.5-turbo

    characteristics associated with charm and physical attractiveness

    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
    Materials
    -0.78
    EMS
    -0.73
     clearance
    -0.73
     Anthem
    -0.72
    mare
    -0.70
     Riot
    -0.68
     overload
    -0.67
     mater
    -0.66
    Gaza
    -0.66
     GEAR
    -0.65
    POSITIVE LOGITS
     charismatic
    2.34
     handsome
    2.19
     charming
    1.72
     narcissistic
    1.64
     charisma
    1.58
     eccentric
    1.51
     manipulative
    1.47
     blond
    1.45
     arrogant
    1.42
     witty
    1.39
    Activations Density 0.073%

    No Known Activations