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. 5624
    Prev
    Next
    INDEX
    Explanations

    terms related to disciplinary actions and penalties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    phrases related to disciplinary actions or penalties

    oai_token-act-pair · gpt-3.5-turbo
    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
    ebus
    -0.82
    sonian
    -0.78
    Transform
    -0.76
    quart
    -0.74
    ertility
    -0.73
    TPP
    -0.70
    cock
    -0.70
    transform
    -0.70
    craft
    -0.69
    Craft
    -0.67
    POSITIVE LOGITS
    ishment
    1.07
     notices
    1.00
     levied
    0.98
     citation
    0.97
     disqual
    0.96
     revoked
    0.95
     probation
    0.93
     punishable
    0.90
     issued
    0.90
     summons
    0.90
    Activations Density 0.179%

    No Known Activations