© 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. 0-RES-JB
    11. 4980
    Prev
    Next
    INDEX
    Explanations

    mentions of specific computer processors, particularly Snapdragon processors

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of processors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Urban
    -0.71
    Urban
    -0.70
     Negro
    -0.69
     Dem
    -0.69
    fell
    -0.67
     Rent
    -0.67
     Gent
    -0.66
    vered
    -0.65
     spots
    -0.65
     Higgins
    -0.65
    POSITIVE LOGITS
     processor
    3.70
     processors
    3.17
     Processor
    2.46
    processor
    2.43
     processing
    1.83
     CPU
    1.73
     chipset
    1.71
    processing
    1.61
     CPUs
    1.57
    Process
    1.49
    Activations Density 0.022%

    No Known Activations