© 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. 1-RES-JB
    6. 11606
    Prev
    Next
    INDEX
    Explanations

    sequences of characters with variable lengths and patterns

    oai_token-act-pair · gpt-3.5-turbo

    repetitions of specific sequences or symbols, potentially representing emotional or emphasis markers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    bender
    -0.66
    cery
    -0.66
     Shah
    -0.65
     CPR
    -0.62
    amus
    -0.61
     Stam
    -0.61
    ullah
    -0.60
    ARY
    -0.60
    OUND
    -0.59
     mash
    -0.59
    POSITIVE LOGITS
    claimed
    0.83
    avorite
    0.80
    interstitial
    0.71
    ettings
    0.70
    uner
    0.70
    Rated
    0.65
    ModLoader
    0.64
    ventus
    0.64
     guiName
    0.62
    VERS
    0.62
    Activations Density 0.000%

    No Known Activations