© 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. 8256
    Prev
    Next
    INDEX
    Explanations

    The neuron is looking for references to spending money

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "spend" in various contexts related to expenditure

    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
     Hazard
    -0.70
    iron
    -0.70
     Verse
    -0.65
    der
    -0.64
     Voice
    -0.64
     Cox
    -0.64
     Asylum
    -0.60
     Rohingya
    -0.60
     reversed
    -0.60
     Relief
    -0.60
    POSITIVE LOGITS
     spend
    3.42
     spends
    2.42
     Spend
    2.36
     spending
    2.15
     spent
    2.08
     Spending
    1.96
     invest
    1.58
     expenditure
    1.49
     dedicate
    1.48
     expenditures
    1.43
    Activations Density 0.010%

    No Known Activations