Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 5492
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to wizards and magical elements

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of a "wizard," particularly in the context of fantasy literature and related themes

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

    No Comments

    Negative Logits
    PER
    -0.82
    uters
    -0.78
     sqor
    -0.75
    tera
    -0.74
    rontal
    -0.71
    76561
    -0.68
    andro
    -0.67
    DonaldTrump
    -0.67
    USE
    -0.66
    ¿½
    -0.65
    POSITIVE LOGITS
     wizard
    0.98
    ry
    0.95
    sonian
    0.94
     Wizard
    0.86
    esses
    0.82
    izard
    0.81
    stru
    0.81
    loo
    0.79
    ery
    0.77
    eers
    0.74
    Activations Density 0.011%

    No Known Activations