© 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. 8-RES-JB
    6. 2073
    Prev
    Next
    INDEX
    Explanations

    references to being aboard a vehicle or transportation-related contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to actions involving transportation or travel incidents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •single token firings
    Fires pretty strongly on " aboard" and less-so on " boarded, "onboard", " veh", " tanker", " hijacked".

    No Comments

    Negative Logits
     Diss
    -0.76
    utenberg
    -0.73
    女
    -0.73
    wu
    -0.73
     pores
    -0.72
    ozo
    -0.71
     ppm
    -0.70
     Discipline
    -0.70
     Lauder
    -0.68
    ipher
    -0.67
    POSITIVE LOGITS
     vehicles
    1.58
     cars
    1.57
     vehicle
    1.54
     taxis
    1.47
     passenger
    1.42
     sedan
    1.39
     automobiles
    1.39
     automobile
    1.38
    wagen
    1.35
     Vehicle
    1.35
    Activations Density 0.983%

    No Known Activations