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. 4944
    Prev
    Next
    INDEX
    Explanations

    polymer-related terms, specifically related to polystyrene

    oai_token-act-pair · gpt-3.5-turbo

    terms related to styling or aesthetics, particularly in materials or artistic contexts

    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
    ajor
    -0.86
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.78
    女
    -0.76
     WARN
    -0.75
    itarian
    -0.70
    ãĥīãĥ©ãĤ´ãĥ³
    -0.68
     Butterfly
    -0.67
    士
    -0.66
     Mutual
    -0.65
     visitation
    -0.64
    POSITIVE LOGITS
    sty
    1.29
    led
    1.03
    les
    0.95
    gian
    0.94
    lish
    0.90
    rene
    0.89
    rette
    0.84
    ling
    0.83
    rous
    0.83
    cil
    0.80
    Activations Density 0.017%

    No Known Activations