Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 3234
    Prev
    Next
    INDEX
    Explanations

    time-related expressions, particularly referring to time periods like "the first quarter."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of time periods, specifically the word "first" in relation to sequential months or quarters

    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
    borgh
    -0.74
    ffen
    -0.65
    tics
    -0.60
    ovi
    -0.58
    atsu
    -0.56
    ingen
    -0.56
     Chips
    -0.55
    ierrez
    -0.55
    Downloadha
    -0.55
    cham
    -0.55
    POSITIVE LOGITS
     half
    1.30
     quarter
    1.19
    half
    1.16
     halves
    1.01
     decade
    1.00
     phase
    0.96
     six
    0.95
     few
    0.95
     week
    0.93
     month
    0.93
    Activations Density 0.076%

    No Known Activations