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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 11-RES-JB
    6. 15247
    Prev
    Next
    INDEX
    Explanations

    phrases related to summoning and sacrificing monsters in a card game

    oai_token-act-pair · gpt-3.5-turbo

    references to card game mechanics involving graveyards and monster summoning

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     blat
    -0.75
    iatus
    -0.71
    gee
    -0.70
     cs
    -0.65
    esm
    -0.65
     todd
    -0.64
     circumcision
    -0.64
    iland
    -0.64
     Lima
    -0.64
     eas
    -0.63
    POSITIVE LOGITS
     Fusion
    0.95
     CARD
    0.93
     Summon
    0.91
     Pend
    0.88
    ributed
    0.88
     Spell
    0.88
     banished
    0.87
     ATK
    0.86
    ributes
    0.86
     card
    0.85
    Activations Density 0.075%

    No Known Activations