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. 17909
    Prev
    Next
    INDEX
    Explanations

    names or titles related to government officials

    oai_token-act-pair · gpt-3.5-turbo

    references to government officials and their roles

    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
    vent
    -0.73
    feat
    -0.70
     Reincarn
    -0.64
     Samurai
    -0.62
    vantage
    -0.61
     torch
    -0.60
    DIS
    -0.59
     lightsaber
    -0.58
     rive
    -0.57
     Totem
    -0.57
    POSITIVE LOGITS
    ial
    1.15
    arians
    1.08
    ially
    1.04
    ials
    0.98
    arian
    0.97
    onse
    0.88
    itol
    0.84
    ional
    0.81
    iture
    0.80
    isms
    0.79
    Activations Density 0.030%

    No Known Activations