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. 5965
    Prev
    Next
    INDEX
    Explanations

    words related to resting or relaxation

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "rest" and its variations

    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
     Curve
    -0.76
     dated
    -0.64
     balloons
    -0.63
     eh
    -0.60
     bytes
    -0.58
     firsthand
    -0.58
     NI
    -0.58
    MK
    -0.57
     MK
    -0.57
     Lilith
    -0.56
    POSITIVE LOGITS
    rest
    4.66
    Rest
    2.17
    restrial
    1.60
     Rest
    1.30
    res
    1.25
    rel
    1.24
    resp
    1.23
    ret
    1.19
    rem
    1.18
    rent
    1.15
    Activations Density 0.006%

    No Known Activations