© 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. 11-RES-JB
    11. 20646
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "week" and related temporal references

    oai_token-act-pair · gpt-3.5-turbo

    mentions of time, specifically the word "week."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    erate
    -0.70
     predec
    -0.64
    ventus
    -0.62
    vae
    -0.60
     plurality
    -0.58
     Journals
    -0.58
    ry
    -0.58
     sync
    -0.57
    gery
    -0.56
     fixation
    -0.55
    POSITIVE LOGITS
     afternoon
    0.91
     evening
    0.80
     alone
    0.78
    night
    0.77
     morning
    0.74
     marked
    0.68
     aboard
    0.68
     amid
    0.66
     marking
    0.66
    flower
    0.66
    Activations Density 0.116%

    No Known Activations