© 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. 2654
    Prev
    Next
    INDEX
    Explanations

    the word "these" followed by a number, highlighting a trend or characteristic of a particular time period or group

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "these days."

    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
    intel
    -0.63
    Pub
    -0.62
    dash
    -0.60
    erate
    -0.59
    iggurat
    -0.59
    reath
    -0.59
    tar
    -0.58
    athon
    -0.58
    ribune
    -0.56
     Catch
    -0.55
    POSITIVE LOGITS
     days
    1.83
    days
    1.24
    Days
    1.14
     Days
    1.12
     nights
    1.12
     mornings
    1.09
     DAY
    1.03
     weeks
    1.02
     times
    1.00
     evenings
    0.95
    Activations Density 0.041%

    No Known Activations