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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 0-RES-JB
    11. 435
    Prev
    Next
    INDEX
    Explanations

    references to physical windows

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "window" in various contexts

    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
     Samoa
    -0.72
    uren
    -0.72
    inos
    -0.70
    Äĵ
    -0.67
    ETH
    -0.67
    eth
    -0.66
    stan
    -0.66
    Ä«
    -0.66
    cipl
    -0.62
    ihil
    -0.62
    POSITIVE LOGITS
     window
    3.83
     windows
    2.82
     Window
    2.62
    window
    2.54
    Window
    1.96
    windows
    1.54
     doorway
    1.49
     door
    1.46
     windshield
    1.45
     pane
    1.45
    Activations Density 0.010%

    No Known Activations