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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 4351
    Prev
    Next
    INDEX
    Explanations

    mentions of various types of craft-related activities, such as crafting, craft beer, and craftsmanship

    oai_token-act-pair · gpt-3.5-turbo

    references to craft-related items or activities, particularly in the context of beer

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Fas
    -1.01
     NCT
    -0.87
     Mum
    -0.84
     Mub
    -0.77
     Tut
    -0.72
     GOODMAN
    -0.72
     Authorities
    -0.71
     Amen
    -0.70
     Burk
    -0.70
     Kenyan
    -0.69
    POSITIVE LOGITS
    smanship
    2.18
    spe
    1.08
    manship
    1.08
    craft
    0.90
    shape
    0.89
     craft
    0.89
    raft
    0.89
     crafts
    0.88
    ivism
    0.88
    rys
    0.86
    Activations Density 0.016%

    No Known Activations