© 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. 17274
    Prev
    Next
    INDEX
    Explanations

    the Cyrillic alphabet letter "Ð" in the text

    oai_token-act-pair · gpt-3.5-turbo

    characters or symbols from a non-Latin script, possibly indicating text encoding issues

    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
     Starr
    -0.70
     Bain
    -0.70
    theless
    -0.69
     Downing
    -0.69
    20439
    -0.69
     Bethlehem
    -0.67
    anwhile
    -0.67
     guiActiveUnfocused
    -0.67
    wagen
    -0.67
    aday
    -0.65
    POSITIVE LOGITS
    ¸
    1.33
    Ķ
    1.30
    ļ
    1.29
    ł
    1.24
    ¾
    1.23
    ¼
    1.23
    ¡
    1.22
    ij
    1.21
    ²
    1.21
    IJ
    1.21
    Activations Density 0.007%

    No Known Activations