© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 12-RES-JB
    11. 4089
    Prev
    Next
    INDEX
    Explanations

    phrases about the challenges of maintaining relationships or bonds between characters in a show or program

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    words that describe qualities or states of being

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    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
    minster
    -0.63
     Kos
    -0.62
     Ig
    -0.61
     Tik
    -0.59
     Scorp
    -0.59
     abundantly
    -0.59
     contribut
    -0.55
     Flo
    -0.54
     Finn
    -0.53
     Mov
    -0.53
    POSITIVE LOGITS
    hement
    1.23
    eness
    1.22
    ilty
    1.06
    ences
    0.96
    ity
    0.96
    cies
    0.94
    ence
    0.93
    ciation
    0.92
    osity
    0.91
    ety
    0.90
    Activations Density 0.116%

    No Known Activations