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. 8-RES-JB
    6. 16141
    Prev
    Next
    INDEX
    Explanations

    adjectives describing emotional states like "frustrating," "tough," and "scary."

    oai_token-act-pair · gpt-3.5-turbo

    emotions and descriptions of challenging and overwhelming situations

    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
    inders
    -0.69
    apons
    -0.69
    mbudsman
    -0.67
    iverpool
    -0.67
    riter
    -0.67
    ournal
    -0.65
    leases
    -0.65
    rongh
    -0.64
    umbn
    -0.64
    ithe
    -0.62
    POSITIVE LOGITS
     uphill
    0.88
     :(
    0.84
    !!
    0.82
    !!!!!
    0.79
    !!!
    0.76
    !!!!
    0.75
     when
    0.74
    ðŁĺ
    0.73
     ðŁĻĤ
    0.73
     smelling
    0.73
    Activations Density 0.211%

    No Known Activations