© 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. 0-RES-JB
    11. 9847
    Prev
    Next
    INDEX
    Explanations

    specific mentions of physical spaces like locations or parts of a building

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of floors in different 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
    ussia
    -0.86
    went
    -0.73
    POST
    -0.68
     Journalism
    -0.67
    ussian
    -0.66
    ils
    -0.63
    download
    -0.61
     Nathaniel
    -0.61
     Saga
    -0.60
    lishes
    -0.60
    POSITIVE LOGITS
     floor
    3.95
     Floor
    3.11
     floors
    2.94
    floor
    2.92
     ceiling
    2.22
     ceilings
    1.72
     roof
    1.49
     carpet
    1.42
     Flo
    1.27
     hallway
    1.24
    Activations Density 0.010%

    No Known Activations