© 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. 1-RES-JB
    11. 855
    Prev
    Next
    INDEX
    Explanations

    terms related to nuclear weapons and nuclear energy

    oai_token-act-pair · gpt-3.5-turbo

    references to nuclear activity or issues

    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
    ean
    -0.79
    regon
    -0.77
    hon
    -0.76
    raised
    -0.75
    anza
    -0.74
    clair
    -0.73
    lay
    -0.72
    odore
    -0.71
    ̶
    -0.71
    ---------------
    -0.70
    POSITIVE LOGITS
     nuclear
    1.13
     reactors
    1.07
     reactor
    1.07
     deterrent
    0.96
     missile
    0.95
     warheads
    0.94
     submarine
    0.92
     submarines
    0.92
     Nuclear
    0.89
     missiles
    0.88
    Activations Density 0.017%

    No Known Activations