Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 14532
    Prev
    Next
    INDEX
    Explanations

    references to the person Saddam Hussein

    oai_token-act-pair · gpt-3.5-turbo

    references to Saddam Hussein

    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
    yp
    -0.73
     OT
    -0.68
     cap
    -0.68
     mode
    -0.67
     temp
    -0.65
     caps
    -0.65
     solo
    -0.63
     bot
    -0.63
    pi
    -0.63
     float
    -0.62
    POSITIVE LOGITS
     Hussein
    4.19
     Hussain
    2.06
     Saddam
    1.76
     Mahmoud
    1.36
     Mubarak
    1.36
     Hassan
    1.34
    Iraq
    1.34
     Hasan
    1.31
     Khalid
    1.31
     Ibrahim
    1.27
    Activations Density 0.008%

    No Known Activations