© 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. 1-RES-JB
    11. 6474
    Prev
    Next
    INDEX
    Explanations

    words related to mechs or mechanical warfare, especially focusing on the term "MechWarrior" and associated terms

    oai_token-act-pair · gpt-3.5-turbo

    references to different types of mechs in a gaming context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    kt
    -0.72
    lv
    -0.70
    æĢ
    -0.70
     epit
    -0.69
     Vatican
    -0.69
     consec
    -0.67
     perpet
    -0.67
     Innocent
    -0.67
     NX
    -0.66
    00007
    -0.66
    POSITIVE LOGITS
    Mech
    1.13
    romancer
    1.11
     Mech
    1.10
    anical
    1.09
    ategory
    1.05
     mech
    1.03
     srf
    1.01
    acter
    0.97
    anooga
    0.94
    istrate
    0.93
    Activations Density 0.013%

    No Known Activations