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. 0-RES-JB
    6. 21652
    Prev
    Next
    INDEX
    Explanations

    references to different centuries

    oai_token-act-pair · gpt-3.5-turbo

    references to specific centuries

    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
    ouk
    -0.65
     attached
    -0.64
     activ
    -0.61
     Support
    -0.61
    support
    -0.59
     TX
    -0.58
    items
    -0.57
     support
    -0.56
    nit
    -0.56
     Eval
    -0.56
    POSITIVE LOGITS
     century
    3.56
    century
    2.87
     Century
    2.55
     millennium
    2.35
     centuries
    2.17
     decade
    2.01
     millennia
    1.81
     decades
    1.43
    ocene
    1.25
     Millennium
    1.23
    Activations Density 0.018%

    No Known Activations