© 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. 1-RES-JB
    11. 12923
    Prev
    Next
    INDEX
    Explanations

    patterns related to metadata or coding structures

    oai_token-act-pair · gpt-3.5-turbo

    content related to social media links or embedded elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Beir
    -0.78
     Samar
    -0.73
     Khe
    -0.70
     Berry
    -0.70
     Ago
    -0.69
     Seah
    -0.67
     Pug
    -0.65
     Cathy
    -0.65
     Pie
    -0.64
     Sie
    -0.63
    POSITIVE LOGITS
    [/
    1.25
    cffffcc
    1.23
    ...]
    1.14
     [/
    1.05
    ][/
    1.03
    catentry
    1.01
    âĢ¦]
    0.99
    âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦âĢ¦
    0.88
    ËĪ
    0.87
    usr
    0.87
    Activations Density 0.005%

    No Known Activations