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. 0-RES-JB
    6. 10833
    Prev
    Next
    INDEX
    Explanations

    mention of legal sentences or penalties

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sentenced," particularly in legal contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Uni
    -0.67
     AAA
    -0.65
     buck
    -0.63
     Discover
    -0.60
     Britann
    -0.59
     Omni
    -0.59
    iscover
    -0.58
     ground
    -0.56
    mega
    -0.56
    side
    -0.55
    POSITIVE LOGITS
     sentenced
    3.47
     jailed
    1.99
     convicted
    1.91
     sentencing
    1.88
     indicted
    1.72
     fined
    1.71
     acquitted
    1.66
     imprisoned
    1.62
     prosecuted
    1.56
     deported
    1.55
    Activations Density 0.011%

    No Known Activations