© 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. 8-RES-JB
    11. 2446
    Prev
    Next
    INDEX
    Explanations

    numeric values

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating contradictory or conditional statements

    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
    tnc
    -0.80
    natureconservancy
    -0.73
    NetMessage
    -0.68
    ħĭ
    -0.62
    opic
    -0.61
    Synopsis
    -0.61
    Hide
    -0.59
    olitical
    -0.58
     earnest
    -0.58
     deg
    -0.58
    POSITIVE LOGITS
     counting
    0.98
     averages
    0.97
     averaged
    0.91
     accounted
    0.87
     averaging
    0.86
    total
    0.86
    average
    0.85
     totaled
    0.82
     555
    0.80
     5000
    0.79
    Activations Density 0.385%

    No Known Activations