© 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. 3-RES-JB
    11. 9112
    Prev
    Next
    INDEX
    Explanations

    special characters, most likely as part of detecting anomalies or errors in the text rather than looking for specific content

    oai_token-act-pair · gpt-3.5-turbo

    special characters and symbols that often indicate formatting or emphasis in text

    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
     deed
    -0.74
    wig
    -0.72
    wagen
    -0.71
     poisoning
    -0.69
    holm
    -0.67
     velvet
    -0.66
     bonded
    -0.64
    ified
    -0.63
     shroud
    -0.62
     sheer
    -0.62
    POSITIVE LOGITS
    £
    1.06
    ï
    1.02
    Ħ
    0.99
    ¼
    0.97
    ł
    0.97
    Ģ
    0.96
    cffffcc
    0.95
    ĺ
    0.94
    Ķ
    0.94
    ¾
    0.93
    Activations Density 0.006%

    No Known Activations