© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 186
    Prev
    Next
    INDEX
    Explanations

    average or averaging statistics related to various activities

    oai_token-act-pair · gpt-3.5-turbo

    references to statistical averages in sports performance

    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
     Nou
    -0.66
    nut
    -0.65
     Liberation
    -0.64
    PU
    -0.63
    ighter
    -0.61
    abet
    -0.61
    ACTED
    -0.61
    aina
    -0.60
    hair
    -0.60
     Space
    -0.59
    POSITIVE LOGITS
     averages
    0.97
    imates
    0.85
     unbeliev
    0.84
     averaging
    0.79
    nesota
    0.79
     averaged
    0.78
     ILCS
    0.77
    oday
    0.76
     variance
    0.75
     ratios
    0.75
    Activations Density 0.008%

    No Known Activations