© 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. 11-RES-JB
    11. 10016
    Prev
    Next
    INDEX
    Explanations

    occurrences of the phrase "earlier this year"

    oai_token-act-pair · gpt-3.5-turbo

    references to the current time, specifically the phrase "this year" or "this month"

    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
     Petra
    -0.66
    sers
    -0.65
    Downloadha
    -0.61
    adle
    -0.61
    ModLoader
    -0.60
    arthy
    -0.58
     HMS
    -0.58
     Abyss
    -0.58
    Gra
    -0.57
     Vict
    -0.56
    POSITIVE LOGITS
     month
    1.82
     year
    1.69
     week
    1.61
     summer
    1.38
     decade
    1.34
     semester
    1.26
     afternoon
    1.25
     offseason
    1.22
     season
    1.20
     morning
    1.20
    Activations Density 0.043%

    No Known Activations