© 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. 0-RES-JB
    11. 8363
    Prev
    Next
    INDEX
    Explanations

    words related to inspectors and investigations

    oai_token-act-pair · gpt-3.5-turbo

    references to inspectors or inspection-related roles

    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
    lation
    -0.72
     viable
    -0.68
     caloric
    -0.68
     amph
    -0.64
    waves
    -0.63
     hormonal
    -0.63
     sperm
    -0.60
     waves
    -0.59
    inav
    -0.59
    pel
    -0.58
    POSITIVE LOGITS
     Inspector
    4.20
     inspector
    3.62
     Inspect
    1.95
     inspectors
    1.90
     Superintendent
    1.71
    ptroller
    1.59
     superintendent
    1.55
    intendent
    1.42
     Sergeant
    1.33
     Inspection
    1.32
    Activations Density 0.015%

    No Known Activations