© 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. 8-RES-JB
    11. 6418
    Prev
    Next
    INDEX
    Explanations

    the word "resort" followed by a gerund form of a verb

    oai_token-act-pair · gpt-3.5-turbo

    mentions of resorts or resort-related activities

    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
    Not in Any Lists

    No Comments

    Negative Logits
    hemat
    -0.79
    rod
    -0.78
    yan
    -0.64
    orate
    -0.63
    mson
    -0.63
    work
    -0.63
    heartedly
    -0.60
    warm
    -0.60
     damn
    -0.59
    paces
    -0.59
    POSITIVE LOGITS
    geoning
    1.02
    itous
    0.78
     resorts
    0.74
     resort
    0.73
    itably
    0.71
    interstitial
    0.71
    pmwiki
    0.69
    ulic
    0.69
    reality
    0.67
    achev
    0.66
    Activations Density 0.057%

    No Known Activations