© 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. 1-RES-JB
    11. 595
    Prev
    Next
    INDEX
    Explanations

    sequences of characters that are not typical words or phrases, but seem to have consistent patterns like special characters and numbers

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of a specific non-standard character or formatting artifact

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ppelin
    -0.71
    enstein
    -0.68
    ENN
    -0.63
     Slayer
    -0.63
    venant
    -0.62
    seless
    -0.62
     Mechdragon
    -0.58
    EVA
    -0.57
    raved
    -0.57
     meal
    -0.56
    POSITIVE LOGITS
    âĢ²
    1.25
    entimes
    0.76
    cation
    0.75
    tis
    0.73
    '-
    0.73
    ATIVE
    0.68
    erous
    0.67
    Solution
    0.66
    acion
    0.66
    ional
    0.65
    Activations Density 0.004%

    No Known Activations