© 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. 21473
    Prev
    Next
    INDEX
    Explanations

    verbs related to starting or beginning a journey or task

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "embark" and its variations, indicating the theme of starting journeys or adventures

    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
    gnu
    -0.66
     bin
    -0.63
     overheard
    -0.59
     variance
    -0.57
     Bin
    -0.56
    ggies
    -0.54
     loophole
    -0.54
     median
    -0.53
     observable
    -0.53
     Accuracy
    -0.52
    POSITIVE LOGITS
     upon
    1.15
    antly
    0.98
    eous
    0.85
    uates
    0.85
     Upon
    0.81
    edIn
    0.81
     anew
    0.77
    TAIN
    0.77
    ¯
    0.76
    uate
    0.76
    Activations Density 0.056%

    No Known Activations