© 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. 8-RES-JB
    11. 1245
    Prev
    Next
    INDEX
    Explanations

    terms related to a specific location, possibly a country or region named Antigua

    oai_token-act-pair · gpt-3.5-turbo

    references to specific attributes or qualities related to certain terms or entities, particularly in connection with "Antigua" and "antimatter."

    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
    ĸļ
    -0.81
     cones
    -0.69
     quarters
    -0.64
     spheres
    -0.64
     envelope
    -0.64
    rely
    -0.63
     princ
    -0.62
     buds
    -0.62
     jewels
    -0.60
     veil
    -0.60
    POSITIVE LOGITS
    icrobial
    0.93
    ctica
    0.90
    ravity
    0.81
    overty
    0.77
    acterial
    0.76
    oli
    0.75
    ircraft
    0.75
    itu
    0.72
    agic
    0.72
    sych
    0.72
    Activations Density 0.067%

    No Known Activations