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. 23576
    Prev
    Next
    INDEX
    Explanations

    industrial entities related to manufacturing processes and products

    oai_token-act-pair · gpt-3.5-turbo

    references to manufacturers and their activities

    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
    hip
    -0.71
     Hare
    -0.69
     tense
    -0.69
    mem
    -0.65
    Person
    -0.63
     Prayer
    -0.63
    anse
    -0.63
    bage
    -0.61
    gal
    -0.61
     Shard
    -0.60
    POSITIVE LOGITS
     manufacturers
    3.58
     manufacturer
    3.17
     manufactures
    2.63
     Manufact
    2.13
     makers
    2.05
     Manufacturer
    2.01
    Manufact
    2.00
     suppliers
    1.93
     automakers
    1.92
     manufacturing
    1.83
    Activations Density 0.019%

    No Known Activations