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. 24483
    Prev
    Next
    INDEX
    Explanations

    numbers, specifically the number 45

    oai_token-act-pair · gpt-3.5-turbo

    references to numerical values, particularly those related to statistics or quantities

    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
    •Embedding Space Digit Features

    No Comments

    Negative Logits
    \\\\
    -0.81
    tain
    -0.66
    ãĤ¨ãĥ«
    -0.65
    iasm
    -0.65
    andre
    -0.61
    uld
    -0.60
    ibr
    -0.60
    inventoryQuantity
    -0.59
    ahl
    -0.59
    uce
    -0.58
    POSITIVE LOGITS
     45
    2.92
     44
    2.03
     46
    2.03
     55
    2.01
     35
    1.91
     48
    1.87
     47
    1.86
     43
    1.84
     40
    1.81
     60
    1.78
    Activations Density 0.022%

    No Known Activations