© 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. 8235
    Prev
    Next
    INDEX
    Explanations

    specific names or terms related to geography and sports, particularly focusing on Dak Prescott, a fountain, and Steve Spurrier

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns related to specific locations or individuals

    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
    Dialogue
    -0.85
    ships
    -0.81
    itions
    -0.74
    ship
    -0.73
    ITED
    -0.70
    Reply
    -0.66
    Torrent
    -0.66
    IENT
    -0.65
    eller
    -0.64
    ecast
    -0.64
    POSITIVE LOGITS
    renheit
    0.80
    toe
    0.79
    olate
    0.77
    ç
    0.72
    olson
    0.71
    tera
    0.71
    plings
    0.71
    merga
    0.70
    pas
    0.70
     drafts
    0.69
    Activations Density 0.115%

    No Known Activations