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. 23771
    Prev
    Next
    INDEX
    Explanations

    the name "Mun" along with different suffixes and word combinations

    oai_token-act-pair · gpt-3.5-turbo

    references to the name "Mun" or related variations

    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
     rept
    -0.73
    elled
    -0.70
     «
    -0.70
     reptiles
    -0.68
    reath
    -0.64
    esty
    -0.64
    orders
    -0.63
     snake
    -0.63
    RESULTS
    -0.62
    step
    -0.61
    POSITIVE LOGITS
     Mun
    4.08
     mun
    1.80
    mun
    1.60
     Bun
    1.30
     Mund
    1.13
     Mau
    1.12
     Mü
    1.07
     Municipal
    1.07
     Kun
    1.06
     Hann
    1.05
    Activations Density 0.020%

    No Known Activations