© 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. 3-RES-JB
    11. 23338
    Prev
    Next
    INDEX
    Explanations

    hedges or cautious terms that express concerns or warnings

    oai_token-act-pair · gpt-3.5-turbo

    expressions indicating caution or warnings about potential negative consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cheong
    -0.79
     Swords
    -0.71
    lee
    -0.70
    ird
    -0.69
     Cooperative
    -0.65
    ulative
    -0.64
    sword
    -0.64
    stood
    -0.63
     Congratulations
    -0.63
    tained
    -0.63
    POSITIVE LOGITS
     lest
    1.22
    00200000
    0.92
     detract
    0.85
     avert
    0.80
    othal
    0.77
    ĸļ
    0.76
     apprehens
    0.75
     cens
    0.75
     indict
    0.74
    entimes
    0.74
    Activations Density 0.007%

    No Known Activations