© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 10768
    Prev
    Next
    INDEX
    Explanations

    references to specific locations, particularly "Syracuse," and specific names, especially "Sammy"

    oai_token-act-pair · gpt-3.5-turbo

    references to the location Syracuse and variations of the name Sammy

    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
    growth
    -0.93
    teen
    -0.81
    ãĥĩãĤ£
    -0.80
    cedented
    -0.80
    wagen
    -0.77
    heid
    -0.75
    ebin
    -0.75
    rodu
    -0.75
    yrinth
    -0.74
    eting
    -0.72
    POSITIVE LOGITS
     Crunch
    0.90
    oya
    0.83
     mosqu
    0.72
    aucuses
    0.71
    amus
    0.68
     Harbor
    0.68
     Slug
    0.67
     Brav
    0.67
     Sapp
    0.66
     enthusi
    0.66
    Activations Density 0.016%

    No Known Activations