© 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. 4668
    Prev
    Next
    INDEX
    Explanations

    references to different versions or qualities of a specific product or character

    oai_token-act-pair · gpt-3.5-turbo

    references to the term "Hero" and related products or characters

    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
    aeda
    -0.95
    rup
    -0.91
    ateur
    -0.79
    ware
    -0.79
    iment
    -0.78
    imentary
    -0.77
    ĸļ
    -0.76
    gow
    -0.75
    creen
    -0.73
    ntil
    -0.73
    POSITIVE LOGITS
    Hero
    0.84
    ku
    0.79
     Reborn
    0.75
     Heroes
    0.72
     Acad
    0.71
     Hero
    0.70
    ines
    0.68
     Fighters
    0.68
    Blade
    0.67
    icide
    0.66
    Activations Density 0.015%

    No Known Activations