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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 10864
    Prev
    Next
    INDEX
    Explanations

    phrases indicating decision-making or actions taken based on a specific situation or circumstance

    oai_token-act-pair · gpt-3.5-turbo

    the word "so" and its variations, indicating a conversational or transitional context

    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
    •Causality
    So + thus, therefore

    No Comments

    Negative Logits
     Modes
    -0.63
    Ranked
    -0.62
    Newsletter
    -0.61
    sha
    -0.60
    driving
    -0.60
     Regions
    -0.60
     ®
    -0.60
    issance
    -0.59
    weights
    -0.58
    degree
    -0.56
    POSITIVE LOGITS
    oner
    1.27
    ooo
    1.21
    oooo
    1.20
    bered
    1.15
     glad
    1.07
     yeah
    1.03
    othe
    1.02
    apy
    0.97
    oooooooo
    0.95
    fter
    0.91
    Activations Density 0.086%

    No Known Activations