© 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. 3-RES-JB
    11. 24052
    Prev
    Next
    INDEX
    Explanations

    geographic locations or proper nouns starting with 'Far'

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific individual or character named "Far."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    sburgh
    -0.91
    ettings
    -0.76
    ichick
    -0.72
    vironment
    -0.69
    essee
    -0.67
    ysis
    -0.67
    代
    -0.66
    ortium
    -0.65
     Kinn
    -0.63
    erous
    -0.62
    POSITIVE LOGITS
    ouk
    0.88
    bent
    0.87
    aday
    0.86
    fur
    0.86
    riers
    0.85
    rier
    0.83
    ghan
    0.82
    rug
    0.82
    away
    0.82
    az
    0.81
    Activations Density 0.013%

    No Known Activations