© 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. 3-RES-JB
    11. 23579
    Prev
    Next
    INDEX
    Explanations

    the first name "Erik" and one occurrence of the first name "Jorge"

    oai_token-act-pair · gpt-3.5-turbo

    the name Erik as well as the name Jorge in various contexts

    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
    roads
    -0.73
    birth
    -0.71
    creen
    -0.70
    ãģĤ
    -0.70
    cock
    -0.68
    talk
    -0.67
    Premium
    -0.67
    ODUCT
    -0.66
    body
    -0.66
    talking
    -0.66
    POSITIVE LOGITS
     Erik
    1.04
     Wem
    0.94
    ildo
    0.94
     Spo
    0.92
    kson
    0.90
    ansson
    0.90
     Bry
    0.89
     Hansen
    0.86
    lund
    0.84
    odox
    0.84
    Activations Density 0.009%

    No Known Activations