© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 4431
    Prev
    Next
    INDEX
    Explanations

    mentions of the long past events or situations

    oai_token-act-pair · gpt-3.5-turbo

    references to historical context or legacy

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

    No Comments

    Negative Logits
    aido
    -0.80
    orney
    -0.79
    omsday
    -0.79
    ouk
    -0.77
    pose
    -0.76
    orc
    -0.75
    tan
    -0.75
    ateurs
    -0.74
    uler
    -0.74
    wer
    -0.73
    POSITIVE LOGITS
     history
    0.77
     record
    0.76
     Revision
    0.75
     KH
    0.70
     Timeline
    0.69
     intertwined
    0.69
     bracelet
    0.68
    ISTORY
    0.68
     incarn
    0.68
     revision
    0.67
    Activations Density 0.024%

    No Known Activations