© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 23247
    Prev
    Next
    INDEX
    Explanations

    the word "mit" followed by a number (likely representing a reference or acronym)

    oai_token-act-pair · gpt-3.5-turbo

    the variations of the word "mit" and its usages across different contexts

    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
     tremend
    -0.85
    Ͻ
    -0.65
    ¥ŀ
    -0.64
    lets
    -0.64
     reconc
    -0.61
    accompan
    -0.61
     resil
    -0.61
    ento
    -0.60
     pharmacies
    -0.59
    spr
    -0.59
    POSITIVE LOGITS
    chell
    1.27
    mit
    1.08
    ochond
    0.97
    igation
    0.93
    rique
    0.86
    igated
    0.86
    ters
    0.85
    idia
    0.82
    ropolitan
    0.80
    ovic
    0.79
    Activations Density 0.012%

    No Known Activations