Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 22236
    Prev
    Next
    INDEX
    Explanations

    adjectives and verbs related to fear and horror

    oai_token-act-pair · gpt-3.5-turbo

    instances of the words "terrifying" and "frightening."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •layer 3 - fear

    No Comments

    Negative Logits
    uties
    -0.84
    ributed
    -0.83
    putable
    -0.82
    yard
    -0.81
    cipl
    -0.81
    elman
    -0.77
    ribut
    -0.77
    pel
    -0.76
    inion
    -0.74
    available
    -0.74
    POSITIVE LOGITS
    ly
    1.07
     unbeliev
    0.95
     ordeal
    0.82
    ingly
    0.79
    NESS
    0.78
    crow
    0.77
     adolesc
    0.77
     horrors
    0.76
     reptiles
    0.76
    LY
    0.74
    Activations Density 0.016%

    No Known Activations