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. 16063
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal or official matters

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the term "subject" in various contexts

    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
    alls
    -0.77
    lux
    -0.76
    swing
    -0.68
    bil
    -0.66
     Rally
    -0.65
     Rae
    -0.63
    iny
    -0.63
    ome
    -0.62
    naires
    -0.62
    thouse
    -0.62
    POSITIVE LOGITS
     subject
    3.89
    subject
    2.78
     Subject
    2.47
     subjects
    2.45
    Subject
    2.36
     topic
    2.19
     Subjects
    1.88
     subjected
    1.52
     topics
    1.42
     Topic
    1.15
    Activations Density 0.017%

    No Known Activations