© 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. 0-RES-JB
    11. 18265
    Prev
    Next
    INDEX
    Explanations

    references to fastening objects with bolts

    oai_token-act-pair · gpt-3.5-turbo

    the presence of the word "bolt" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Sidney
    -0.76
    rine
    -0.76
     Lilly
    -0.75
     Sana
    -0.75
    00200000
    -0.74
     Lau
    -0.70
     Pigs
    -0.70
     Riyadh
    -0.70
     Ric
    -0.69
     Saints
    -0.69
    POSITIVE LOGITS
     bolt
    3.27
     bolts
    2.81
     Bolt
    2.40
    bolt
    2.16
     bolted
    1.94
     brace
    1.42
     hob
    1.34
     bol
    1.32
     bree
    1.13
     boulder
    1.12
    Activations Density 0.013%

    No Known Activations