© 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. 10-RES-JB
    11. 6403
    Prev
    Next
    INDEX
    Explanations

    phrases related to cause and effect

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to causation and dependency in complex situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ille
    -0.78
    soever
    -0.74
    iest
    -0.73
    æ³
    -0.73
    monary
    -0.73
    oided
    -0.72
    ogly
    -0.72
    iership
    -0.72
    ura
    -0.72
    erella
    -0.71
    POSITIVE LOGITS
     how
    0.97
     recent
    0.97
     longstanding
    0.94
     reports
    0.87
     persistent
    0.87
     perceptions
    0.87
     the
    0.86
     fears
    0.84
     ongoing
    0.84
     assurances
    0.84
    Activations Density 0.285%

    No Known Activations