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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 17995
    Prev
    Next
    INDEX
    Explanations

    mentions of unauthorized access or breaches of security

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "hacked" and its related forms

    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
     attraction
    -0.76
     Relative
    -0.71
    Ont
    -0.70
     alcohol
    -0.70
     Alcohol
    -0.68
    Duration
    -0.67
     Tanner
    -0.67
     Duration
    -0.66
    onder
    -0.65
     attractions
    -0.65
    POSITIVE LOGITS
     hacked
    3.83
     hacking
    2.01
     hacks
    1.97
     hack
    1.95
     hackers
    1.77
     hacker
    1.70
    hack
    1.62
     hijacked
    1.50
     stolen
    1.48
     leaked
    1.38
    Activations Density 0.013%

    No Known Activations