© 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. 3-RES-JB
    11. 6146
    Prev
    Next
    INDEX
    Explanations

    terms related to the action of developing something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "developed" in various contexts related to creation or innovation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Keeper
    -0.84
    que
    -0.69
    lua
    -0.67
    nea
    -0.64
    zzi
    -0.62
    duty
    -0.61
     Correction
    -0.60
     Horus
    -0.60
    wa
    -0.59
    wered
    -0.59
    POSITIVE LOGITS
    developed
    0.97
     developed
    0.84
    urally
    0.83
    irtual
    0.80
     develops
    0.77
    wcs
    0.77
    WER
    0.76
    iator
    0.75
    rums
    0.74
     Seym
    0.73
    Activations Density 0.022%

    No Known Activations