© 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. 0-RES-JB
    11. 12045
    Prev
    Next
    INDEX
    Explanations

    numerical values related to amounts of money or quantities

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the number "300" or numeric values formatted similarly

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     jur
    -0.65
     interpretation
    -0.65
     Nash
    -0.63
    kell
    -0.63
     recognition
    -0.61
     miss
    -0.61
     Harding
    -0.61
     view
    -0.61
     identification
    -0.60
     afore
    -0.60
    POSITIVE LOGITS
    300
    3.34
    400
    2.41
    600
    2.34
    500
    2.31
    700
    2.26
    800
    2.20
    200
    2.15
    900
    2.11
    250
    1.99
    3000
    1.99
    Activations Density 0.017%

    No Known Activations