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. 3-RES-JB
    6. 8577
    Prev
    Next
    INDEX
    Explanations

    phrases related to emotions or essence

    oai_token-act-pair · gpt-3.5-turbo

    references to "heart" and its metaphorical uses in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     newsp
    -0.83
     Motorsport
    -0.71
    VICE
    -0.68
    ggies
    -0.67
     Sigma
    -0.66
    qus
    -0.64
    dq
    -0.62
     CSI
    -0.61
     Lavrov
    -0.61
     Sessions
    -0.61
    POSITIVE LOGITS
    strings
    1.09
    lake
    0.91
    heart
    0.90
     heart
    0.89
    ily
    0.88
    hole
    0.87
    beat
    0.87
    piece
    0.85
     artery
    0.85
    burn
    0.85
    Activations Density 0.017%

    No Known Activations