© 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. 2549
    Prev
    Next
    INDEX
    Explanations

    words related to scientific experiments or tests

    oai_token-act-pair · gpt-3.5-turbo

    references to clinical trials

    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
    otle
    -0.82
    resent
    -0.73
    afort
    -0.73
    ippi
    -0.72
     Boh
    -0.72
    Security
    -0.69
    DEBUG
    -0.66
    ols
    -0.65
    ips
    -0.65
    achu
    -0.64
    POSITIVE LOGITS
     trials
    0.88
     trial
    0.88
     balloon
    0.88
     Trials
    0.85
     Trial
    0.76
     unsuccessfully
    0.75
     balloons
    0.75
     "$:/
    0.73
    naire
    0.69
    eer
    0.68
    Activations Density 0.012%

    No Known Activations