© 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. 1-RES-JB
    11. 3705
    Prev
    Next
    INDEX
    Explanations

    words related to transitions or sequences of events

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and connecting words that indicate relationships between ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     tremend
    -0.87
     hardest
    -0.75
     referen
    -0.75
     rare
    -0.74
     calming
    -0.73
     warr
    -0.73
     hard
    -0.72
    £ı
    -0.71
     recession
    -0.71
     exha
    -0.71
    POSITIVE LOGITS
    iste
    0.82
    pad
    0.81
    imi
    0.81
    illian
    0.81
    lys
    0.81
    atis
    0.80
    tec
    0.80
    hess
    0.80
    pedia
    0.80
    vier
    0.79
    Activations Density 0.358%

    No Known Activations