© 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. 4-RES-JB
    11. 9148
    Prev
    Next
    INDEX
    Explanations

    mentions of tax-related terms and phrases

    oai_token-act-pair · gpt-3.5-turbo

    terms related to tax cuts and breaks, particularly for the wealthy

    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
    Ô
    -0.84
    Ul
    -0.81
     Nik
    -0.74
     Nag
    -0.71
    Sensor
    -0.69
     Wand
    -0.68
     Voy
    -0.67
    Ship
    -0.66
    Attempts
    -0.66
     Cornel
    -0.64
    POSITIVE LOGITS
    payer
    0.94
     giveaways
    0.91
     amnesty
    0.88
     goodies
    0.87
    amacare
    0.86
    riage
    0.82
     benefiting
    0.81
     giveaway
    0.81
     loopholes
    0.81
     entitle
    0.81
    Activations Density 0.040%

    No Known Activations