© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 10-RES-JB
    6. 8050
    Prev
    Next
    INDEX
    Explanations

    references related to a specific fictional magical school

    oai_token-act-pair · gpt-3.5-turbo

    references to the Harry Potter series and its characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Words related to Harry Potter

    Explanation Uploaded by User
    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
    •Harry Potter
    •Interesting neurons
    The neuron fires around concepts related to Harry Potter but often on the preceding word. I.e., Harry Potter context.

    No Comments

    Negative Logits
    Houston
    -1.06
     $$
    -1.04
    igslist
    -0.94
    unicip
    -0.92
    Dallas
    -0.91
     NASL
    -0.91
    Austin
    -0.90
    $$
    -0.90
    Texas
    -0.88
     MLG
    -0.87
    POSITIVE LOGITS
     Rowling
    1.86
     Hogwarts
    1.81
     Dumbledore
    1.71
     Malfoy
    1.68
     Hermione
    1.67
     Snape
    1.63
     Voldemort
    1.63
     Weasley
    1.60
     Harry
    1.57
     Potter
    1.57
    Activations Density 0.220%

    No Known Activations