© 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. 9524
    Prev
    Next
    INDEX
    Explanations

    mentions of entities with the abbreviation "EC" followed by a number (potentially a rating or classification)

    oai_token-act-pair · gpt-3.5-turbo

    references to the acronym "EC" and its related 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
    pocket
    -0.74
     ç¥ŀ
    -0.72
     Sack
    -0.68
     Heard
    -0.67
     Touch
    -0.65
     Markus
    -0.64
     Kush
    -0.64
    yards
    -0.63
    down
    -0.63
     Grimm
    -0.63
    POSITIVE LOGITS
    ritical
    1.00
    oin
    1.00
    entric
    1.00
    ontent
    0.99
    redits
    0.95
    ERN
    0.95
    ycle
    0.95
    OLOG
    0.94
    zech
    0.94
    EC
    0.94
    Activations Density 0.007%

    No Known Activations