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. 8-RES-JB
    6. 55
    Prev
    Next
    INDEX
    Explanations

    commas in text, particularly in sports or news-related contexts.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @johnny

    commas in sentences related to sports news, particularly soccer/football.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @johnny

    terms related to sports competitions or events

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks, particularly commas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Utah
    -0.79
    sylvania
    -0.77
    center
    -0.76
    meyer
    -0.76
    ¶
    -0.73
     Harbor
    -0.70
    imize
    -0.70
     nonex
    -0.69
     favors
    -0.68
     Louisiana
    -0.68
    POSITIVE LOGITS
     whilst
    1.57
     Whilst
    1.39
     organising
    1.22
     apologise
    1.20
     fue
    1.20
     emphas
    1.19
     recognise
    1.18
     organised
    1.17
     fuelled
    1.17
     organisers
    1.16
    Activations Density 0.539%

    No Known Activations