© 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. 0-RES-JB
    11. 10424
    Prev
    Next
    INDEX
    Explanations

    lines or sequences of text that mention a physical line, such as a police line or a line of defense

    oai_token-act-pair · gpt-3.5-turbo

    phrases referring to "lines."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    borgh
    -0.74
    pec
    -0.73
    bnb
    -0.73
     Nasa
    -0.72
    adobe
    -0.70
    iac
    -0.67
    abouts
    -0.67
    prison
    -0.67
    apist
    -0.65
    vere
    -0.65
    POSITIVE LOGITS
     line
    3.81
     lines
    2.83
    line
    2.66
     Line
    2.65
    Line
    2.44
     LINE
    2.25
    lines
    2.16
     Lines
    2.15
    LINE
    1.91
    liner
    1.44
    Activations Density 0.019%

    No Known Activations