© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 15775
    Prev
    Next
    INDEX
    Explanations

    references to different types of boxes, especially when related to physical objects or tools

    oai_token-act-pair · gpt-3.5-turbo

    references to boxes

    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
     suicides
    -0.68
     migr
    -0.68
    ustainable
    -0.68
     Thousand
    -0.66
     fertil
    -0.66
     fertility
    -0.64
     sympath
    -0.63
     Seraph
    -0.62
     tresp
    -0.62
     Mour
    -0.62
    POSITIVE LOGITS
    box
    4.40
    boxes
    3.46
    Box
    2.57
    boxing
    2.52
    BOX
    2.39
     box
    2.12
     Box
    2.01
     boxes
    1.79
     BOX
    1.74
    kit
    1.37
    Activations Density 0.007%

    No Known Activations