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. 246
    Prev
    Next
    INDEX
    Explanations

    mentions of floor coverings, specifically "carpet" and "rug"

    oai_token-act-pair · gpt-3.5-turbo

    terms related to carpets and rugs

    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
    edin
    -0.79
     Stru
    -0.74
     Gon
    -0.70
     SA
    -0.68
     Ath
    -0.67
    sv
    -0.65
     Odin
    -0.64
    illation
    -0.64
    Birth
    -0.64
    Ol
    -0.64
    POSITIVE LOGITS
     carpet
    4.03
     rug
    2.49
     carp
    2.08
     curtains
    1.36
     sofa
    1.34
     mattress
    1.33
     mats
    1.23
     couch
    1.23
     floor
    1.21
     blanket
    1.18
    Activations Density 0.011%

    No Known Activations