© 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. 2240
    Prev
    Next
    INDEX
    Explanations

    instances where the text discusses different groups of people or individuals, often with contrasting characteristics or actions

    oai_token-act-pair · gpt-3.5-turbo

    references to groups of people and their actions or beliefs in a political or social context

    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
    estones
    -0.75
    estone
    -0.68
    ĸļ
    -0.66
    ongo
    -0.61
    lly
    -0.60
    ovember
    -0.59
    VERTISEMENT
    -0.58
     Retrieved
    -0.57
    ogan
    -0.57
    itu
    -0.56
    POSITIVE LOGITS
    who
    1.96
     who
    1.92
     whom
    1.68
    whose
    1.55
     whose
    1.54
    Who
    1.42
     Who
    1.38
     WHO
    1.07
    WHO
    1.02
    omever
    1.00
    Activations Density 0.218%

    No Known Activations