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. 0-RES-JB
    6. 16108
    Prev
    Next
    INDEX
    Explanations

    references to various styles, activities, and aspects related to dancing

    oai_token-act-pair · gpt-3.5-turbo

    references to the act of dancing

    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
    Hub
    -0.68
    LR
    -0.67
    insured
    -0.66
    installed
    -0.64
     Checking
    -0.64
     Pack
    -0.64
    ortium
    -0.64
     Detected
    -0.63
     Harrington
    -0.63
    chens
    -0.63
    POSITIVE LOGITS
     dance
    3.65
     dances
    2.77
     Dance
    2.68
     dancing
    2.37
     danced
    2.30
     dancers
    2.15
     dancer
    2.14
     ballet
    1.98
     Dancing
    1.72
     disco
    1.54
    Activations Density 0.011%

    No Known Activations