© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 19950
    Prev
    Next
    INDEX
    Explanations

    instances of the word "last" occurring in the text

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating finality or closure, particularly in relation to significant events or circumstances

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

    No Comments

    Negative Logits
    arth
    -0.76
    hovah
    -0.72
    skirts
    -0.72
     oneself
    -0.69
     themselves
    -0.68
    atis
    -0.68
     itself
    -0.67
    \\\\\\\\\\\\\\\\
    -0.66
     Vish
    -0.65
    rish
    -0.65
    POSITIVE LOGITS
     foray
    0.96
     counterparts
    0.85
     cousin
    0.83
     outing
    0.83
    favorite
    0.78
     counterpart
    0.78
     birthday
    0.77
     comeback
    0.75
    iasm
    0.73
     namesake
    0.73
    Activations Density 0.185%

    No Known Activations