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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 10-RES-JB
    11. 2277
    Prev
    Next
    INDEX
    Explanations

    phrases indicating that it's too early to make a specific determination or prediction

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating timeliness or the nascent stage of events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ractor
    -0.61
     racket
    -0.58
    bryce
    -0.56
    severe
    -0.55
    rac
    -0.53
    Tur
    -0.53
    viol
    -0.53
     LIFE
    -0.53
    avier
    -0.53
     spac
    -0.53
    POSITIVE LOGITS
    ittal
    0.79
     estimation
    0.75
     infancy
    0.74
     stages
    0.69
    eral
    0.65
     approximation
    0.65
     forecasting
    0.65
    verified
    0.64
    gradation
    0.63
    2019
    0.63
    Activations Density 0.094%

    No Known Activations