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

    Neuronpedia

    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. 1-RES-JB
    11. 22580
    Prev
    Next
    INDEX
    Explanations

    words related to summoning or invoking

    oai_token-act-pair · gpt-3.5-turbo

    words associated with magical or supernatural conjurations

    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
    OPLE
    -0.73
    TPS
    -0.66
     Carnegie
    -0.65
     incentive
    -0.64
     Preferred
    -0.62
    aimon
    -0.61
     Hide
    -0.60
    ESA
    -0.59
    mm
    -0.59
     Nano
    -0.59
    POSITIVE LOGITS
     conj
    1.32
    unct
    1.09
    ugal
    1.07
    uncture
    1.03
    uration
    0.93
    oked
    0.89
    oint
    0.83
     exorc
    0.82
     Conj
    0.81
    raph
    0.80
    Activations Density 0.010%

    No Known Activations