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. 8803
    Prev
    Next
    INDEX
    Explanations

    information related to legal sentences and serving time

    oai_token-act-pair · gpt-3.5-turbo

    references to serving time or prison sentences

    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
     Collider
    -0.67
     stumbling
    -0.66
    TPPStreamerBot
    -0.64
     tackling
    -0.61
    duc
    -0.61
    Jump
    -0.58
    Org
    -0.58
    aq
    -0.58
     embryos
    -0.57
    Winged
    -0.57
    POSITIVE LOGITS
     faithfully
    0.90
    entials
    0.75
    olit
    0.74
    andestine
    0.73
    umbn
    0.72
     subpoena
    0.71
     constituents
    0.69
    umbnail
    0.69
    udeb
    0.69
    olesterol
    0.68
    Activations Density 0.108%

    No Known Activations