© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 22312
    Prev
    Next
    INDEX
    Explanations

    references to the term "Commons" in the context of licenses or permissions

    oai_token-act-pair · gpt-3.5-turbo

    references to "Commons," likely relating to governmental or public discussions and legislation

    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
     caliber
    -0.66
     exceeded
    -0.65
     tang
    -0.65
    alo
    -0.65
    phthal
    -0.64
    gran
    -0.64
    caliber
    -0.63
    hur
    -0.63
     surfing
    -0.63
    isher
    -0.62
    POSITIVE LOGITS
     Commons
    4.47
     commons
    2.54
     Lords
    1.66
     Parliamentary
    1.28
     Houses
    1.16
     Parliament
    1.15
     parliamentary
    1.09
     MPs
    1.07
    Comm
    1.04
     Westminster
    1.04
    Activations Density 0.017%

    No Known Activations