© 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. 4-RES-JB
    11. 4784
    Prev
    Next
    INDEX
    Explanations

    mentions of the country Australia

    oai_token-act-pair · gpt-3.5-turbo

    references to Australia

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

    No Comments

    Negative Logits
    outer
    -0.76
    osen
    -0.72
    acco
    -0.71
    Lens
    -0.70
    abet
    -0.69
    ows
    -0.69
    ser
    -0.69
    soDeliveryDate
    -0.68
    angler
    -0.68
    Jump
    -0.66
    POSITIVE LOGITS
     Australia
    1.00
     Defence
    0.85
    Australia
    0.83
     Broadcasting
    0.82
     Australians
    0.81
     Aboriginal
    0.80
     Tasmania
    0.79
     Airlines
    0.79
     Geographic
    0.78
    gdala
    0.77
    Activations Density 0.013%

    No Known Activations