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. 4-RES-JB
    6. 23312
    Prev
    Next
    INDEX
    Explanations

    references to figures of authority, possibly with religious associations, such as fathers or priests

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "Father" in a religious or familial context

    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
    yrinth
    -0.73
    ucket
    -0.72
    ellen
    -0.72
    AW
    -0.71
    Tokens
    -0.70
    NRS
    -0.68
    ELL
    -0.68
     Flavoring
    -0.64
    rawdownloadcloneembedreportprint
    -0.63
    eers
    -0.63
    POSITIVE LOGITS
    hood
    1.03
     Father
    0.88
    angelo
    0.86
     Almighty
    0.84
    Father
    0.83
    heses
    0.80
    hesis
    0.80
     Abraham
    0.77
    father
    0.76
     Fathers
    0.72
    Activations Density 0.009%

    No Known Activations