© 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. 8-RES-JB
    11. 7676
    Prev
    Next
    INDEX
    Explanations

    proper names, specifically "Reid"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the person Harry Reid

    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
    andise
    -1.03
    omaly
    -0.78
    eatures
    -0.77
    tesy
    -0.72
     ILCS
    -0.71
    ruciating
    -0.70
    teenth
    -0.68
    tips
    -0.66
     dehyd
    -0.63
     behav
    -0.63
    POSITIVE LOGITS
    burgh
    0.86
    ifying
    0.80
    igan
    0.73
     Reid
    0.70
    sburg
    0.70
     Hoffman
    0.70
     Mysteries
    0.68
    uler
    0.68
     Rib
    0.67
    wd
    0.67
    Activations Density 0.030%

    No Known Activations