© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 9593
    Prev
    Next
    INDEX
    Explanations

    tax-related terms, specifically related to deductions and adjustments to income

    oai_token-act-pair · gpt-3.5-turbo

    terms related to tax deductions and financial adjustments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Nieto
    -0.72
    anco
    -0.71
    hell
    -0.71
    cock
    -0.66
     Image
    -0.62
    EEK
    -0.62
    OCK
    -0.61
    oys
    -0.61
     constitu
    -0.61
    igma
    -0.60
    POSITIVE LOGITS
     deductions
    1.17
     deduct
    1.11
     deduction
    1.04
     deducted
    1.03
     deductible
    1.02
    payer
    0.94
     allowance
    0.89
     exemptions
    0.88
     allowable
    0.87
     allowances
    0.85
    Activations Density 0.012%

    No Known Activations