© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    11. 6810
    Prev
    Next
    INDEX
    Explanations

    special characters and symbols used for technical patterns or code, possibly related to programming or mathematical equations

    oai_token-act-pair · gpt-3.5-turbo

    symbols and characters used in code or mathematical expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ividual
    -0.76
    lain
    -0.75
    itia
    -0.73
    oran
    -0.68
     Trilogy
    -0.68
    unky
    -0.66
     Scotia
    -0.64
     Beir
    -0.63
     seiz
    -0.63
     Riot
    -0.62
    POSITIVE LOGITS
    {\
    0.94
    ł
    0.87
    sin
    0.85
    Ĵ
    0.84
    Ī
    0.82
    temp
    0.81
    ¥
    0.79
    ãģ£
    0.79
    ¿
    0.79
    ¤
    0.79
    Activations Density 0.030%

    No Known Activations