Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 10475
    Prev
    Next
    INDEX
    Explanations

    mentions of general life experiences and circumstances

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of life and its various aspects or experiences

    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
    sonian
    -0.90
     Cosponsors
    -0.83
    CAST
    -0.78
    mented
    -0.73
    EStreamFrame
    -0.72
    POS
    -0.69
    IUM
    -0.68
    atchewan
    -0.68
    essee
    -0.68
    agger
    -0.67
    POSITIVE LOGITS
     expectancy
    1.07
    guards
    1.01
    boat
    0.99
    blood
    0.91
    boats
    0.91
    guard
    0.90
    forms
    0.83
     everlasting
    0.79
     raft
    0.79
    liness
    0.77
    Activations Density 0.041%

    No Known Activations