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. 0-RES-JB
    6. 8924
    Prev
    Next
    INDEX
    Explanations

    the word "precious" with high activation

    oai_token-act-pair · gpt-3.5-turbo

    words that denote value, particularly in terms of being cherished or irreplaceable

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Strikes
    -0.71
    ence
    -0.71
    ent
    -0.70
    Mania
    -0.66
    ooters
    -0.65
    off
    -0.62
    itch
    -0.62
     Strike
    -0.62
    ment
    -0.62
    Off
    -0.62
    POSITIVE LOGITS
     precious
    3.58
     priceless
    1.94
     valuable
    1.73
     scarce
    1.48
     dear
    1.47
     dearly
    1.46
     prized
    1.42
     fragile
    1.39
     sacred
    1.33
     cherished
    1.32
    Activations Density 0.015%

    No Known Activations