© 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. 4-RES-JB
    11. 739
    Prev
    Next
    INDEX
    Explanations

    references to specific names like "Gwyn" and "Oswalt."

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of specific names, particularly "Wes" and "Oswalt"

    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
    unin
    -0.91
     locality
    -0.69
    avez
    -0.68
    ional
    -0.67
    onom
    -0.66
    omsky
    -0.65
    rero
    -0.64
     predicting
    -0.62
    onomic
    -0.62
     predictive
    -0.62
    POSITIVE LOGITS
    wyn
    1.23
    er
    0.87
    burn
    0.86
    ner
    0.84
    æ©Ł
    0.82
    robe
    0.82
    ת
    0.81
    igans
    0.79
    ¯¯
    0.76
    bridge
    0.75
    Activations Density 0.006%

    No Known Activations