© 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. 2-RES-JB
    11. 424
    Prev
    Next
    INDEX
    Explanations

    the sequence of letters "ne"

    oai_token-act-pair · gpt-4-turbo

    the word "ne" followed by a single-letter code

    oai_token-act-pair · gpt-3.5-turbo

    the occurrence of the sequence "ne"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rador
    -0.84
    Reviewer
    -0.78
    DOWN
    -0.74
     bearer
    -0.73
    DragonMagazine
    -0.73
    displayText
    -0.73
    inarily
    -0.72
    ^^^^
    -0.70
    hips
    -0.69
    ãĥ¼ãĥĨãĤ£
    -0.69
    POSITIVE LOGITS
    arest
    1.09
    lde
    0.93
    braska
    0.93
    gan
    0.90
    volent
    0.89
    ople
    0.88
    theless
    0.86
    cker
    0.84
    issance
    0.84
    igh
    0.84
    Activations Density 0.020%

    No Known Activations