© 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. 0-RES-JB
    11. 5233
    Prev
    Next
    INDEX
    Explanations

    phrases related to military organizations and operations, particularly those involving the term "Joint"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Joint" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     atheism
    -0.80
    advertising
    -0.78
    ãĤ¦ãĤ¹
    -0.73
    hin
    -0.71
     sorcery
    -0.71
     fres
    -0.68
     racism
    -0.67
     save
    -0.67
     erase
    -0.66
    ca
    -0.66
    POSITIVE LOGITS
     Joint
    3.89
     joint
    2.39
     joints
    1.62
     jointly
    1.28
     Combined
    1.12
     Cooperative
    1.09
     JO
    1.00
     Conc
    1.00
    oint
    0.98
     Permanent
    0.93
    Activations Density 0.005%

    No Known Activations