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. 8891
    Prev
    Next
    INDEX
    Explanations

    mentions of a physical box or organization with "Box" in the name

    oai_token-act-pair · gpt-3.5-turbo

    references to "box" or "BOX" 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
     Seraph
    -0.74
    esan
    -0.72
     intel
    -0.67
     HM
    -0.66
     methane
    -0.64
     dracon
    -0.64
     Erin
    -0.64
    ¶ħ
    -0.63
     chant
    -0.62
     Michele
    -0.62
    POSITIVE LOGITS
    Box
    4.15
    BOX
    2.56
    box
    2.45
     Box
    2.42
     BOX
    2.11
    boxing
    1.93
    boxes
    1.91
     box
    1.82
     boxes
    1.59
     boxed
    1.25
    Activations Density 0.012%

    No Known Activations