© 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. 22208
    Prev
    Next
    INDEX
    Explanations

    legal or financial terms related to exploiting loopholes or gaps in regulations

    oai_token-act-pair · gpt-3.5-turbo

    references to legal loopholes and exploitation of such loopholes

    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
     reconstruct
    -0.73
     prem
    -0.64
    acid
    -0.64
     precipitation
    -0.64
     radi
    -0.61
    ior
    -0.61
    aud
    -0.60
    itect
    -0.59
    ald
    -0.59
     elevation
    -0.59
    POSITIVE LOGITS
     loophole
    3.83
     loopholes
    3.74
     backdoor
    1.49
     amnesty
    1.25
     exemptions
    1.15
     circumvent
    1.12
     cheating
    1.02
     workaround
    1.00
     glitches
    0.98
     looph
    0.98
    Activations Density 0.019%

    No Known Activations