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. 22114
    Prev
    Next
    INDEX
    Explanations

    phrases related to various types of accidents or disasters

    oai_token-act-pair · gpt-3.5-turbo

    instances and variations of the word "crash."

    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
    ival
    -0.72
    advertising
    -0.64
     consideration
    -0.61
    LGBT
    -0.59
    utenberg
    -0.59
     Aqu
    -0.59
    ivalent
    -0.59
     allot
    -0.58
    imaru
    -0.58
     considering
    -0.58
    POSITIVE LOGITS
     crashed
    3.60
     crashes
    2.41
     crashing
    2.28
     crash
    2.13
     collided
    2.12
    Crash
    1.77
     wrecked
    1.71
     Crash
    1.68
     collapsed
    1.59
     downed
    1.46
    Activations Density 0.013%

    No Known Activations