© 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. 20223
    Prev
    Next
    INDEX
    Explanations

    phrases related to something being prominent or significant within a text

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "heavily" indicating significant intensity or emphasis in contexts

    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
    vity
    -0.86
    iry
    -0.77
    arium
    -0.74
    anasia
    -0.69
    absolute
    -0.69
     Liberation
    -0.68
    eln
    -0.68
    ottage
    -0.66
    ridge
    -0.66
    antics
    -0.66
    POSITIVE LOGITS
     reliant
    1.07
     influenced
    1.04
     traff
    1.00
     weighted
    0.99
     invested
    0.96
     indebted
    0.96
     taxed
    0.96
     impacted
    0.95
     relied
    0.94
     subsidized
    0.94
    Activations Density 0.035%

    No Known Activations