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. 4-RES-JB
    6. 3460
    Prev
    Next
    INDEX
    Explanations

    terms related to romantic relationships

    oai_token-act-pair · gpt-3.5-turbo

    mentions of romantic themes and relationships

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

    No Comments

    Negative Logits
    upon
    -0.98
    avis
    -0.91
    Downloadha
    -0.89
    */(
    -0.87
    ldon
    -0.80
    ulhu
    -0.78
    redd
    -0.78
    ridges
    -0.77
    paio
    -0.76
    etts
    -0.76
    POSITIVE LOGITS
     romantic
    0.92
    ized
    0.89
    istic
    0.85
    ization
    0.83
    ties
    0.82
     monog
    0.81
    izing
    0.80
    istically
    0.76
     romance
    0.73
     comed
    0.73
    Activations Density 0.009%

    No Known Activations