© 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 EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 3445
    Prev
    Next
    INDEX
    Explanations

    abstract concepts and philosophical ideas related to reason, knowledge, power, and virtue

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to reason and rational thought

    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
     remotely
    -0.61
    Enlarge
    -0.59
     printer
    -0.59
    iera
    -0.58
     rainy
    -0.56
    ELL
    -0.56
    sit
    -0.55
     Lena
    -0.54
     pocket
    -0.54
     Paddock
    -0.54
    POSITIVE LOGITS
    lessness
    1.38
    fulness
    1.30
    uality
    1.17
    liness
    1.16
    cellence
    1.05
    ativity
    1.05
    ocracy
    1.03
    ivism
    1.03
    ignty
    0.98
    iveness
    0.96
    Activations Density 0.370%

    No Known Activations