© 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. 2-RES-JB
    11. 12864
    Prev
    Next
    INDEX
    Explanations

    references to specific names of people or characters

    oai_token-act-pair · gpt-3.5-turbo

    references to characters related to Willow and Spike from the Buffy the Vampire Slayer series

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rontal
    -0.91
    rals
    -0.87
    orical
    -0.87
    orically
    -0.85
    ribution
    -0.83
    yrinth
    -0.80
    ifax
    -0.79
    apolis
    -0.78
    ogether
    -0.78
    juven
    -0.77
    POSITIVE LOGITS
     Summers
    1.06
     Buffy
    0.97
     Willow
    0.94
     Spike
    0.89
     Xander
    0.87
     Giles
    0.83
     McGee
    0.82
     Angel
    0.81
     Whedon
    0.81
     Rosenberg
    0.80
    Activations Density 0.015%

    No Known Activations