© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 473
    Prev
    Next
    INDEX
    Explanations

    mentions of the character Batman

    oai_token-act-pair · gpt-4-turbo

    references to and mentions of Batman

    oai_token-act-pair · gpt-3.5-turbo

    references to Batman and related media

    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
    mble
    -1.15
    lopp
    -0.86
    ateur
    -0.82
    rition
    -0.74
    ensable
    -0.74
    vant
    -0.73
    ateurs
    -0.73
    ced
    -0.72
    ormons
    -0.72
    hower
    -0.72
    POSITIVE LOGITS
     Begins
    1.00
     Arkham
    0.96
    Batman
    0.95
     Batman
    0.91
     Returns
    0.75
     Superman
    0.75
     Bros
    0.74
     Nolan
    0.71
     villains
    0.70
     Ez
    0.69
    Activations Density 0.009%

    No Known Activations