© 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 EvalsExportsAPI 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. 3-RES-JB
    11. 12728
    Prev
    Next
    INDEX
    Explanations

    instances where the word "handful" is used

    oai_token-act-pair · gpt-3.5-turbo

    mentions of a small number of items or examples

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Train
    -0.68
    Interstitial
    -0.68
    mediate
    -0.67
    abolic
    -0.66
    agnar
    -0.66
    hemat
    -0.65
    Adv
    -0.64
    structed
    -0.63
     illum
    -0.63
     embr
    -0.62
    POSITIVE LOGITS
    eenth
    0.90
    een
    0.88
     dozen
    0.86
     handful
    0.78
    dozen
    0.77
     4090
    0.73
    ortun
    0.73
    ãĤ©
    0.72
     thousand
    0.71
     ILCS
    0.70
    Activations Density 0.011%

    No Known Activations