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. 20925
    Prev
    Next
    INDEX
    Explanations

    words related to law enforcement actions like raids and crackdowns

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "raids" and its derivatives related to military or law enforcement actions

    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
     Springer
    -0.78
     Einstein
    -0.71
    erved
    -0.69
    Nik
    -0.69
    vec
    -0.68
    Ped
    -0.68
    CV
    -0.67
     Distance
    -0.66
     elect
    -0.66
    puter
    -0.66
    POSITIVE LOGITS
     raids
    3.78
     raid
    3.13
     raiding
    2.87
     raided
    2.83
     Raid
    2.01
    Raid
    1.93
    raid
    1.53
     SWAT
    1.29
     swoop
    1.28
     crackdown
    1.28
    Activations Density 0.021%

    No Known Activations