© 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. 0-RES-JB
    11. 2096
    Prev
    Next
    INDEX
    Explanations

    characters or symbols in a specific non-Latin alphabet

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of a specific character or symbol

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     succ
    -0.67
     rain
    -0.66
     diffusion
    -0.65
     bond
    -0.64
     figure
    -0.62
     distance
    -0.60
     smugg
    -0.60
     multiply
    -0.60
     thanks
    -0.59
     Berlin
    -0.59
    POSITIVE LOGITS
    Į
    4.54
    į
    2.05
    IJ
    1.99
    İ
    1.91
    Ī
    1.88
    ı
    1.87
    ħ
    1.81
    ĭ
    1.78
    ī
    1.77
    ¬
    1.71
    Activations Density 0.004%

    No Known Activations