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. 7693
    Prev
    Next
    INDEX
    Explanations

    phrases related to the time needed to accomplish a task or reach a goal

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing the concepts of effort and duration

    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
    tor
    -0.75
    luaj
    -0.71
    tions
    -0.68
    holm
    -0.67
    lib
    -0.65
     Owl
    -0.64
    tx
    -0.63
     Preview
    -0.62
    dor
    -0.61
    raid
    -0.61
    POSITIVE LOGITS
     aback
    0.88
    aways
    0.85
     advantage
    0.80
    ynthesis
    0.76
    YR
    0.76
     place
    0.71
    umes
    0.71
     precedence
    0.71
    ioxide
    0.70
     precautions
    0.70
    Activations Density 0.077%

    No Known Activations