© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 3-RES-JB
    11. 4538
    Prev
    Next
    INDEX
    Explanations

    mentions of the country "Yemen."

    oai_token-act-pair · gpt-3.5-turbo

    references to Yemen

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ablishment
    -0.92
    orsche
    -0.77
    ainer
    -0.76
    ciation
    -0.73
    âķIJâķIJ
    -0.71
     DRM
    -0.70
    ++++++++++++++++
    -0.69
    ivities
    -0.67
     LAPD
    -0.66
    gling
    -0.64
    POSITIVE LOGITS
    halla
    0.81
    ites
    0.80
    beans
    0.76
     Yemen
    0.76
    bean
    0.73
    esque
    0.70
    ican
    0.70
     Arabia
    0.70
    atar
    0.70
     Houth
    0.66
    Activations Density 0.003%

    No Known Activations