© 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. 1-RES-JB
    11. 23319
    Prev
    Next
    INDEX
    Explanations

    references related to the magical world of Hogwarts and magical tools like wands and brooms

    oai_token-act-pair · gpt-3.5-turbo

    references to the Harry Potter universe, particularly elements related to Hogwarts, wands, and prominent characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Fathers
    -0.70
    TAG
    -0.67
     Samar
    -0.63
    etitive
    -0.63
    ï¸ı
    -0.62
    anos
    -0.62
     Rivers
    -0.61
    iets
    -0.60
    anke
    -0.60
    Ocean
    -0.59
    POSITIVE LOGITS
     Dumbledore
    1.04
     Hermione
    1.00
     Voldemort
    0.92
     Weasley
    0.90
    umbledore
    0.86
     Hogwarts
    0.86
     robes
    0.86
     Snape
    0.82
    uggle
    0.80
     appar
    0.78
    Activations Density 0.030%

    No Known Activations