© 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. 4-RES-JB
    11. 16838
    Prev
    Next
    INDEX
    Explanations

    terms related to actions and concepts in different contexts

    oai_token-act-pair · gpt-3.5-turbo

    significant nouns and terms related to various processes and entities in scientific, legal, and structural contexts

    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
     FANTASY
    -0.75
    姫
    -0.74
    worldly
    -0.66
    Weather
    -0.66
    cipled
    -0.64
    poon
    -0.56
    hiro
    -0.56
    å°Ĩ
    -0.56
     Helpful
    -0.56
    concert
    -0.56
    POSITIVE LOGITS
    eers
    0.81
     itself
    0.79
    osphere
    0.71
     debacle
    0.70
     fiasco
    0.70
    ariat
    0.69
    atorium
    0.69
    aires
    0.68
     bandwagon
    0.67
    ocalypse
    0.62
    Activations Density 0.847%

    No Known Activations