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. 8801
    Prev
    Next
    INDEX
    Explanations

    words related to opposition or negation

    oai_token-act-pair · gpt-3.5-turbo

    words and concepts related to antisocial behavior or attitudes

    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
     Duchess
    -0.78
     Falls
    -0.75
     Dynamics
    -0.70
     Warfare
    -0.67
    Transaction
    -0.67
    IRO
    -0.67
     Chiefs
    -0.66
     Penet
    -0.66
     Bundy
    -0.66
     Glacier
    -0.64
    POSITIVE LOGITS
    pace
    1.22
    paces
    1.19
    ocial
    1.17
    earch
    1.07
    terday
    1.00
    chool
    0.98
    chwitz
    0.98
    peed
    0.97
    leep
    0.97
    cript
    0.97
    Activations Density 0.028%

    No Known Activations