© 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. 10-RES-JB
    11. 8584
    Prev
    Next
    INDEX
    Explanations

    instances of colons indicating explanations or lists

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to reasoning and justification

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ascus
    -0.76
    thur
    -0.74
    apsed
    -0.72
    agraph
    -0.67
    rez
    -0.67
    é¾
    -0.66
    Fuck
    -0.66
    emort
    -0.66
    ocide
    -0.66
    cember
    -0.65
    POSITIVE LOGITS
     reducing
    1.00
     it
    0.97
     reduces
    0.95
     facilitating
    0.92
     lowering
    0.89
     lowers
    0.88
     increased
    0.88
     increases
    0.86
     they
    0.86
     eliminating
    0.86
    Activations Density 0.337%

    No Known Activations