Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 4-RES-JB
    6. 21937
    Prev
    Next
    INDEX
    Explanations

    actions related to crafting and creating objects

    oai_token-act-pair · gpt-3.5-turbo

    terms related to tools and instructions for DIY projects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Samar
    -0.90
    ONSORED
    -0.83
     Fatal
    -0.79
     subscribing
    -0.76
     Americas
    -0.73
     Incarn
    -0.73
     Paran
    -0.73
     Ronaldo
    -0.71
     Niger
    -0.68
     Leap
    -0.67
    POSITIVE LOGITS
     adhesive
    0.97
     thickness
    0.95
     acrylic
    0.93
     glue
    0.89
     arrang
    0.88
     sty
    0.87
     srfAttach
    0.86
     crochet
    0.86
     silicone
    0.86
     insulation
    0.84
    Activations Density 0.299%

    No Known Activations