© 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. 18648
    Prev
    Next
    INDEX
    Explanations

    I'm sorry, but based on the provided activations, it is difficult to clearly summarize what Neuron 4 is looking for. There is a variety of seemingly random text content, making it challenging to pinpoint a specific pattern or theme that the neuron is primarily activating for

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the website "Infowars."

    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
     Rolls
    -0.68
     bye
    -0.66
     Dunham
    -0.62
     uninterrupted
    -0.62
     bom
    -0.61
     valiant
    -0.59
     Winston
    -0.58
     successors
    -0.58
     davidjl
    -0.57
     Hug
    -0.57
    POSITIVE LOGITS
    owship
    1.01
    cial
    0.99
    ructure
    0.97
    iband
    0.97
    renheit
    0.95
    essim
    0.90
    ruct
    0.89
    urated
    0.81
    aneous
    0.79
    luster
    0.79
    Activations Density 0.036%

    No Known Activations