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. 4-RES-JB
    6. 5631
    Prev
    Next
    INDEX
    Explanations

    references to groups of people or individuals

    oai_token-act-pair · gpt-3.5-turbo

    references to groups of people and social relationships

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

    No Comments

    Negative Logits
     Accessed
    -0.79
    DEN
    -0.76
     EDT
    -0.63
     NOAA
    -0.62
     COUNTY
    -0.62
    ashington
    -0.61
     NET
    -0.61
    inational
    -0.60
    WASHINGTON
    -0.60
     Pentagon
    -0.59
    POSITIVE LOGITS
    emate
    0.84
     etiquette
    0.83
    ickle
    0.81
    erker
    0.80
    folk
    0.80
    girl
    0.79
    etter
    0.77
    friends
    0.75
    girls
    0.74
    Ĭ±
    0.74
    Activations Density 0.342%

    No Known Activations