© 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. 7067
    Prev
    Next
    INDEX
    Explanations

    narrative descriptions of actions, events, and interactions between individuals

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of actions and events involving digital communication

    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
    uably
    -0.67
     Historically
    -0.57
     guiName
    -0.55
     globally
    -0.52
    pires
    -0.52
    omics
    -0.51
    busters
    -0.51
    rosso
    -0.51
    osponsors
    -0.51
     geographically
    -0.50
    POSITIVE LOGITS
    anus
    0.60
     hurried
    0.60
     startled
    0.56
    RESULTS
    0.55
     frantic
    0.54
     filthy
    0.54
     unsuspecting
    0.53
     unconscious
    0.52
     exha
    0.52
     horrified
    0.51
    Activations Density 2.551%

    No Known Activations