© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 8-RES-JB
    11. 2650
    Prev
    Next
    INDEX
    Explanations

    the number "dozen" occurring as part of a count

    oai_token-act-pair · gpt-3.5-turbo

    references to quantities, specifically the word "dozen" and its variations

    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
    estine
    -0.78
    bro
    -0.76
    ICO
    -0.73
    usher
    -0.71
    ainer
    -0.71
    earch
    -0.69
    iner
    -0.68
    claimed
    -0.65
    hra
    -0.65
    inth
    -0.63
    POSITIVE LOGITS
     thousand
    1.12
     dozen
    0.97
    uously
    0.91
     hundred
    0.90
     Thousand
    0.89
     consecutive
    0.87
    uous
    0.85
    acity
    0.84
    fold
    0.82
     different
    0.80
    Activations Density 0.043%

    No Known Activations