© 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. 14649
    Prev
    Next
    INDEX
    Explanations

    phrases starting with special characters and ending with numbers, possibly related to citations or formatting

    oai_token-act-pair · gpt-3.5-turbo

    symbols or characters that indicate a specific format or notation

    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
    ©¶æ
    -0.70
    inctions
    -0.65
     Kills
    -0.65
    igan
    -0.64
    ²¾
    -0.61
     Capture
    -0.60
     Plastic
    -0.59
    quished
    -0.59
     Able
    -0.57
    £ı
    -0.57
    POSITIVE LOGITS
    cause
    0.83
    LP
    0.78
    said
    0.75
    QB
    0.75
    especially
    0.73
    Zip
    0.71
    fortunately
    0.70
    )'
    0.69
     preval
    0.66
    according
    0.66
    Activations Density 0.341%

    No Known Activations