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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 2923
    Prev
    Next
    INDEX
    Explanations

    references to Wikipedia

    oai_token-act-pair · gpt-3.5-turbo

    mentions of Wikipedia and related references

    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
     uncond
    -0.69
    atch
    -0.65
     pressed
    -0.65
    hea
    -0.64
     festive
    -0.61
    erc
    -0.60
     tal
    -0.60
     beads
    -0.59
    ebted
    -0.59
    icip
    -0.59
    POSITIVE LOGITS
     Wikipedia
    3.87
    Wikipedia
    3.27
     wik
    2.57
     Wikimedia
    2.52
    wikipedia
    2.51
     Wik
    2.05
    ipedia
    1.99
     Wiki
    1.98
     wiki
    1.92
    Wik
    1.91
    Activations Density 0.020%

    No Known Activations