© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 18812
    Prev
    Next
    INDEX
    Explanations

    events or situations where something is physically held or scheduled

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "held" indicating actions or events that involve possession or organization

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bm
    -0.63
    onal
    -0.63
    amorph
    -0.62
    eme
    -0.60
    elo
    -0.58
    eto
    -0.57
     Diesel
    -0.57
    oxide
    -0.57
    iates
    -0.57
     havoc
    -0.57
    POSITIVE LOGITS
     held
    3.56
     Held
    2.21
    held
    2.17
     holds
    1.69
     hold
    1.63
     holding
    1.49
     convened
    1.40
     Holding
    1.37
    Hold
    1.36
     hosted
    1.35
    Activations Density 0.020%

    No Known Activations