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. 3501
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "ec" followed by a single-digit or a two-digit number, which may indicate references to specific contexts or entities

    oai_token-act-pair · gpt-3.5-turbo

    references to various religious groups or communities

    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
    MENTS
    -0.74
     Archdemon
    -0.69
     Desire
    -0.69
     Ou
    -0.67
     Angry
    -0.64
     Ashe
    -0.64
    FTWARE
    -0.64
     Rapp
    -0.63
     Maker
    -0.62
     Sorce
    -0.62
    POSITIVE LOGITS
    stasy
    1.26
    uador
    1.23
    lectic
    1.17
    ogl
    1.14
    utive
    1.11
    inct
    1.06
    urrent
    1.02
    oder
    1.01
    rypt
    0.99
    rypted
    0.98
    Activations Density 0.012%

    No Known Activations