© 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. 8-RES-JB
    11. 7004
    Prev
    Next
    INDEX
    Explanations

    phrases related to jurisdiction or regulation

    oai_token-act-pair · gpt-3.5-turbo

    references to specific locations or jurisdictions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    wen
    -0.70
    caster
    -0.68
    wo
    -0.67
    ãĥ»
    -0.63
    eton
    -0.61
    oka
    -0.61
    eness
    -0.61
    \'
    -0.59
    illon
    -0.58
     lately
    -0.58
    POSITIVE LOGITS
     same
    1.29
     highest
    1.09
     cheapest
    1.05
     utmost
    1.03
     lowest
    1.03
     entirety
    1.00
    same
    0.99
     safest
    0.99
     confines
    0.98
     fullest
    0.96
    Activations Density 0.452%

    No Known Activations