© 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. 20023
    Prev
    Next
    INDEX
    Explanations

    phrases related to the concept of using something, particularly in a regulatory or controlled context

    oai_token-act-pair · gpt-3.5-turbo

    phrases referring to the concept of usage or implementation of various items or technologies

    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
    ellow
    -0.70
     Mous
    -0.67
    Ń·
    -0.66
     Barron
    -0.62
     Bert
    -0.62
     Empires
    -0.61
    ¶ħ
    -0.61
     Barg
    -0.60
     Emir
    -0.60
     Airlines
    -0.60
    POSITIVE LOGITS
    fulness
    1.10
    fully
    0.99
    ability
    0.98
    FUL
    0.95
    itures
    0.87
    ancies
    0.85
    ables
    0.84
    ivation
    0.83
    abilities
    0.82
    nels
    0.81
    Activations Density 0.077%

    No Known Activations