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. 6106
    Prev
    Next
    INDEX
    Explanations

    the word "whilst" in sentences

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "whilst" and related expressions of time or conditional scenarios in a formal context

    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
     ÂŃ
    -1.37
     —
    -0.85
     postwar
    -0.82
    ÂŃ
    -0.80
     Web
    -0.76
     ''
    -0.76
     nond
    -0.75
     advisers
    -0.74
     counterterrorism
    -0.73
    —"
    -0.73
    POSITIVE LOGITS
     whilst
    1.88
     Whilst
    1.82
     alot
    1.49
     Youtube
    1.37
     independ
    1.36
     doesnt
    1.34
     thats
    1.33
     didnt
    1.32
    Firstly
    1.29
     secondly
    1.29
    Activations Density 0.137%

    No Known Activations