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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 13889
    Prev
    Next
    INDEX
    Explanations

    descriptions of car accidents

    oai_token-act-pair · gpt-3.5-turbo

    incidents involving vehicle collisions and accidents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     interrogation
    -0.77
     persuasion
    -0.74
     disinfect
    -0.73
     Guantanamo
    -0.72
     interrog
    -0.72
    HCR
    -0.72
    isSpecialOrderable
    -0.71
    CHQ
    -0.71
    Tea
    -0.71
    ħĭ
    -0.69
    POSITIVE LOGITS
     collided
    1.65
     collision
    1.59
     crashes
    1.47
     crash
    1.45
     collisions
    1.42
     crashed
    1.39
     driver
    1.25
     collide
    1.21
     bicycl
    1.21
     derail
    1.19
    Activations Density 0.278%

    No Known Activations