© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 8-RES-JB
    11. 11140
    Prev
    Next
    INDEX
    Explanations

    combinations of words related to choices or actions, often presented in the form of questions

    oai_token-act-pair · gpt-3.5-turbo

    questions that contrast choices or options

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    sworth
    -0.70
    tips
    -0.66
    checked
    -0.54
    holding
    -0.54
     Letter
    -0.54
    ISON
    -0.53
    ream
    -0.53
     Pen
    -0.52
     gratification
    -0.52
    Pen
    -0.51
    POSITIVE LOGITS
    姫
    0.92
    abouts
    0.90
     simply
    0.85
    phans
    0.83
     outright
    0.82
     alternatively
    0.81
    éŃĶ
    0.80
     merely
    0.79
    depending
    0.74
    ifacts
    0.73
    Activations Density 0.173%

    No Known Activations