© 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. 2-RES-JB
    11. 342
    Prev
    Next
    INDEX
    Explanations

    phrases denoting duration or time lengths

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "lasted" denoting the duration of time an event or situation persisted

    oai_token-act-pair · gpt-4-turbo

    instances of the word "last" in various forms, indicating duration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    eers
    -0.75
    ledged
    -0.72
    raised
    -0.72
    ledge
    -0.67
    erning
    -0.67
    upon
    -0.66
    rim
    -0.66
    afort
    -0.65
    oration
    -0.64
    irlfriend
    -0.63
    POSITIVE LOGITS
     lasted
    1.00
     lasting
    0.96
     destro
    0.87
    ĸļ
    0.87
     therap
    0.85
     leng
    0.80
     lasts
    0.78
     longer
    0.78
     Gleaming
    0.76
     slee
    0.76
    Activations Density 0.008%

    No Known Activations