© 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. 7-RES-JB
    11. 201
    Prev
    Next
    INDEX
    Explanations

    phrases related to conspiracy or illegal collaboration between individuals

    oai_token-act-pair · gpt-3.5-turbo

    references to consular activities or roles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     negatives
    -0.74
     oats
    -0.74
    shire
    -0.71
    neath
    -0.70
     Welsh
    -0.69
     Hour
    -0.69
     Grail
    -0.68
    lihood
    -0.68
     Waters
    -0.68
     Abyss
    -0.68
    POSITIVE LOGITS
    oled
    1.21
    cription
    1.13
    cript
    1.11
    oling
    1.01
    igned
    1.00
    cient
    0.98
    cientious
    0.96
    ignment
    0.96
    ented
    0.94
    ort
    0.94
    Activations Density 0.027%

    No Known Activations