© 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. 2364
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to buildings or locations

    oai_token-act-pair · gpt-3.5-turbo

    references to "Trump Tower."

    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
    dit
    -0.78
    é»Ĵ
    -0.77
    milo
    -0.72
    cause
    -0.71
    ãģ®éŃĶ
    -0.68
    sync
    -0.66
    conscious
    -0.65
    hiro
    -0.65
    ppo
    -0.65
    stract
    -0.63
    POSITIVE LOGITS
     Towers
    1.15
     Tower
    1.08
     skysc
    0.98
    tower
    0.96
     towers
    0.93
     tower
    0.91
     Heist
    0.81
     balcony
    0.80
    fall
    0.80
     Bridge
    0.80
    Activations Density 0.011%

    No Known Activations