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. 17147
    Prev
    Next
    INDEX
    Explanations

    references to times in Pacific Time (PT)

    oai_token-act-pair · gpt-3.5-turbo

    references to specific time zones or periods

    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
    castle
    -0.79
    orah
    -0.73
     grate
    -0.69
    aum
    -0.68
    azard
    -0.65
     Heller
    -0.65
     Lauder
    -0.65
     settled
    -0.64
    love
    -0.63
     Castle
    -0.62
    POSITIVE LOGITS
     PT
    3.93
    PT
    2.13
     MT
    1.78
     ET
    1.67
     PST
    1.46
     CT
    1.43
     CST
    1.35
     GP
    1.27
     KP
    1.25
     IST
    1.24
    Activations Density 0.012%

    No Known Activations