© 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. 8-RES-JB
    11. 10956
    Prev
    Next
    INDEX
    Explanations

    proper nouns and associated descriptors such as names, brands, and titles

    oai_token-act-pair · gpt-3.5-turbo

    names of individuals or entities associated with achievements or recognition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    icol
    -0.75
    osa
    -0.72
     Gutenberg
    -0.72
    Newsletter
    -0.70
    hack
    -0.68
    alin
    -0.67
    )|
    -0.67
    oller
    -0.66
    olog
    -0.66
    Correction
    -0.65
    POSITIVE LOGITS
     nevertheless
    1.11
     nonetheless
    1.07
     meticulously
    0.91
     poised
    0.89
     tirelessly
    0.86
     boldly
    0.85
     irresist
    0.85
     unmist
    0.83
     effortlessly
    0.83
     relentlessly
    0.83
    Activations Density 0.528%

    No Known Activations