© 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 EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 20821
    Prev
    Next
    INDEX
    Explanations

    phrases related to accidents and emergencies

    oai_token-act-pair · gpt-3.5-turbo

    objects or elements that convey danger or a hazardous situation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     chapters
    -1.01
     equivalents
    -0.94
    ernels
    -0.94
     adaptations
    -0.90
     modes
    -0.89
     virtues
    -0.89
     formats
    -0.89
     personalities
    -0.89
    inations
    -0.89
     Languages
    -0.88
    POSITIVE LOGITS
     truck
    1.22
     fence
    1.22
     crane
    1.21
     tractor
    1.20
     wagon
    1.19
     billboard
    1.19
     boulder
    1.17
     statue
    1.11
     SUV
    1.07
     mailbox
    1.06
    Activations Density 0.354%

    No Known Activations