© 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. 1-RES-JB
    11. 6767
    Prev
    Next
    INDEX
    Explanations

    words related to fear or danger

    oai_token-act-pair · gpt-3.5-turbo

    expressions of fear or something that is perceived as frightening

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •layer 1 - fear

    No Comments

    Negative Logits
    ribute
    -0.90
    ributes
    -0.89
    ribut
    -0.88
    elf
    -0.85
    blance
    -0.85
    iciency
    -0.84
    ributed
    -0.84
    inion
    -0.78
    ceptions
    -0.78
    ribution
    -0.77
    POSITIVE LOGITS
     scary
    1.07
    crow
    1.02
     clown
    0.89
     scare
    0.78
     unbeliev
    0.77
     nightmare
    0.75
     shenan
    0.73
     Jagu
    0.71
     scares
    0.71
     suspic
    0.71
    Activations Density 0.005%

    No Known Activations