© 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. 4-RES-JB
    11. 10467
    Prev
    Next
    INDEX
    Explanations

    phrases related to events or actions that occurred before a specific point in time

    oai_token-act-pair · gpt-3.5-turbo

    the word "prior" indicating previous events or actions

    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
    aden
    -0.78
    rosso
    -0.72
    RO
    -0.71
    asp
    -0.70
    ickle
    -0.65
    Baby
    -0.64
    %%%%
    -0.63
    beans
    -0.63
    tower
    -0.63
    girls
    -0.62
    POSITIVE LOGITS
    itiz
    1.26
    itized
    1.08
    etheless
    1.04
    ities
    0.97
    ebin
    0.85
    emort
    0.78
     generations
    0.78
    icip
    0.77
    cies
    0.77
     authenticated
    0.71
    Activations Density 0.015%

    No Known Activations