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. 8806
    Prev
    Next
    INDEX
    Explanations

    mentions of plane hijacking events

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the term "hijack" and its variations

    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
     Scotch
    -0.74
    inel
    -0.70
    VG
    -0.69
    rated
    -0.68
    avis
    -0.66
    EX
    -0.66
    arts
    -0.66
    Arc
    -0.65
    tern
    -0.65
     Raven
    -0.65
    POSITIVE LOGITS
     hij
    4.06
     hijacked
    2.53
     spoof
    1.31
     hijab
    1.27
     kidn
    1.23
     sabotage
    1.15
     Haj
    1.07
     disgu
    1.06
     derail
    1.06
     kidnap
    1.02
    Activations Density 0.017%

    No Known Activations