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. 3063
    Prev
    Next
    INDEX
    Explanations

    mentions of the country Yemen

    oai_token-act-pair · gpt-3.5-turbo

    references to Yemen and its geopolitical situation

    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
     PI
    -0.72
    linger
    -0.65
     posit
    -0.65
    Pi
    -0.64
    PI
    -0.63
    Frame
    -0.63
     cones
    -0.63
     photon
    -0.62
     BT
    -0.61
    bolt
    -0.60
    POSITIVE LOGITS
     Yemen
    3.87
     Yemeni
    3.06
     Houth
    2.46
     Sana
    2.10
     Somalia
    2.06
     Aden
    1.83
     Afghanistan
    1.73
     Sudan
    1.70
     Libya
    1.70
     Oman
    1.69
    Activations Density 0.012%

    No Known Activations