© 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. 10-RES-JB
    11. 23368
    Prev
    Next
    INDEX
    Explanations

    specific technical terms or programming-related keywords

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns and names related to entertainment, organizations, and specific concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ensical
    -0.65
    itious
    -0.60
    risome
    -0.58
     seismic
    -0.57
     speculation
    -0.55
    idious
    -0.54
     Entered
    -0.52
     kinds
    -0.51
     comparisons
    -0.51
     sorts
    -0.49
    POSITIVE LOGITS
     itself
    0.78
    stice
    0.72
    reciation
    0.65
    Ń·
    0.64
     Advocate
    0.64
    swers
    0.64
    brids
    0.64
    edia
    0.63
    Enh
    0.63
    iets
    0.62
    Activations Density 0.648%

    No Known Activations