© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 24519
    Prev
    Next
    INDEX
    Explanations

    computer programming code snippets or related text sequences containing specific characters

    oai_token-act-pair · gpt-3.5-turbo

    repetitive patterns or symbols that are likely encoding specific formats or characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Mous
    -0.85
     scrambling
    -0.67
     Belg
    -0.66
    theless
    -0.66
     ESA
    -0.66
     shroud
    -0.64
     Triangle
    -0.64
     Faw
    -0.63
     Ascension
    -0.63
     Archangel
    -0.60
    POSITIVE LOGITS
    ľ
    1.57
    ª
    1.53
    ¤
    1.50
    ¡
    1.50
    ł
    1.47
    ¦
    1.46
    ¼
    1.45
    Ķ
    1.44
    ´
    1.42
    ¢
    1.41
    Activations Density 0.157%

    No Known Activations