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

    Neuronpedia

    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. 2196
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to the location "Dunedin."

    oai_token-act-pair · gpt-3.5-turbo

    references to the name "Dun" in various contexts

    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
    OHN
    -0.81
    NRS
    -0.74
    âĸĪâĸĪ
    -0.70
    xual
    -0.69
    Magikarp
    -0.66
     BDS
    -0.64
     Pyro
    -0.61
    ocrats
    -0.61
    aceous
    -0.61
    76561
    -0.60
    POSITIVE LOGITS
    wich
    1.03
    lap
    0.97
    querque
    0.96
    edin
    0.95
    leys
    0.92
    wind
    0.92
    igan
    0.91
    ogh
    0.91
    geon
    0.89
    agon
    0.89
    Activations Density 0.005%

    No Known Activations