© 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. 9528
    Prev
    Next
    INDEX
    Explanations

    words or patterns related to mixed characters and symbols, possibly indicating an issue with character encoding or decoding

    oai_token-act-pair · gpt-3.5-turbo

    Arabic characters

    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
     Cly
    -0.80
    onge
    -0.77
    arching
    -0.74
    estern
    -0.73
    iko
    -0.73
     Wilmington
    -0.69
    erton
    -0.68
    tten
    -0.68
     Pyth
    -0.68
     Myster
    -0.67
    POSITIVE LOGITS
    Ù
    1.63
    ÙĨ
    1.58
    Ùĩ
    1.57
    ا
    1.54
    د
    1.54
    Ø
    1.50
    اØ
    1.50
    Ùħ
    1.48
    ر
    1.44
    ت
    1.44
    Activations Density 0.002%

    No Known Activations