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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 0-RES-JB
    11. 23110
    Prev
    Next
    INDEX
    Explanations

    words related to collisions or accidents

    oai_token-act-pair · gpt-3.5-turbo

    references to collisions, particularly with high activation on individual incidents and related discussions

    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
    endas
    -0.70
     Um
    -0.66
    isites
    -0.65
     Tuls
    -0.65
    otent
    -0.64
    nuts
    -0.64
     Agenda
    -0.63
     Ost
    -0.63
     Offer
    -0.62
    tm
    -0.62
    POSITIVE LOGITS
     collision
    3.86
     collisions
    3.16
     collide
    2.28
     collided
    2.20
     crash
    1.59
    coll
    1.43
     accident
    1.43
     clash
    1.36
     crashes
    1.32
    Coll
    1.31
    Activations Density 0.019%

    No Known Activations