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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 7-RES-JB
    6. 5632
    Prev
    Next
    INDEX
    Explanations

    phrases related to superlatives or descriptions of things as the "world's most" or "world's largest."

    oai_token-act-pair · gpt-3.5-turbo

    the possessive form of the word "world."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Lauder
    -0.80
    yang
    -0.67
     ..............
    -0.66
    nery
    -0.65
    docker
    -0.62
    ertodd
    -0.61
     Ital
    -0.61
    Å¡
    -0.61
    powder
    -0.60
    Wan
    -0.60
    POSITIVE LOGITS
     tallest
    0.97
     premiere
    0.87
     foremost
    0.82
     largest
    0.82
     oldest
    0.82
    selves
    0.81
     Greatest
    0.81
     longest
    0.80
     favourite
    0.79
     smallest
    0.78
    Activations Density 0.096%

    No Known Activations