© 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. 20872
    Prev
    Next
    INDEX
    Explanations

    numbers representing quantities or figures

    oai_token-act-pair · gpt-3.5-turbo

    numerical values, especially those related to amounts and specifications

    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
    ouri
    -0.87
    ored
    -0.81
    bara
    -0.80
    alid
    -0.80
    andering
    -0.78
    gotten
    -0.78
    cho
    -0.78
    rosso
    -0.77
    arter
    -0.76
    alde
    -0.75
    POSITIVE LOGITS
    mAh
    1.19
     MHz
    0.90
     NOTICE
    0.87
    MHz
    0.86
     certs
    0.85
     rpm
    0.85
    8000
    0.84
    é¾įåĸļ士
    0.79
     Hz
    0.76
     msec
    0.75
    Activations Density 0.019%

    No Known Activations