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. 11-RES-JB
    6. 10355
    Prev
    Next
    INDEX
    Explanations

    phrases related to forgiveness and religious teachings

    oai_token-act-pair · gpt-3.5-turbo

    references to dialogue and discussions involving religious or moral questions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    abase
    -0.78
    ormal
    -0.69
    ussions
    -0.69
     strutConnector
    -0.68
    olar
    -0.64
    ettings
    -0.64
    details
    -0.63
    GM
    -0.62
    binary
    -0.62
    tsky
    -0.61
    POSITIVE LOGITS
     thou
    1.56
     Thou
    1.48
     thy
    1.31
     ye
    1.30
     Thy
    1.22
     Blessed
    1.14
     thee
    1.12
     unto
    1.11
     shalt
    1.10
    joice
    1.08
    Activations Density 0.154%

    No Known Activations