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. 8-RES-JB
    6. 22229
    Prev
    Next
    INDEX
    Explanations

    mentions of "commitment" or "commission" in different contexts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "comm" or its variations, indicating a focus on communication or commitment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     vulgar
    -0.68
    lings
    -0.67
    chest
    -0.65
    worms
    -0.62
    vae
    -0.62
    geon
    -0.61
    loo
    -0.60
     Balk
    -0.60
    lv
    -0.59
     ultras
    -0.58
    POSITIVE LOGITS
    ittee
    1.31
    onsense
    1.26
    issions
    1.20
    ittal
    1.17
    ittees
    1.17
    mercial
    1.16
    encing
    1.14
    utation
    1.13
    ences
    1.13
    otion
    1.13
    Activations Density 0.025%

    No Known Activations