© 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. 11-RES-JB
    11. 2722
    Prev
    Next
    INDEX
    Explanations

    phrases related to future plans or intentions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "be" in various grammatical forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    encies
    -0.76
    pmwiki
    -0.74
    uyomi
    -0.74
    atures
    -0.68
    rencies
    -0.68
    pedia
    -0.67
     Measures
    -0.65
    bits
    -0.65
     SetTextColor
    -0.63
    yrinth
    -0.63
    POSITIVE LOGITS
     pissed
    1.12
     punished
    1.05
     kidding
    1.03
     deported
    1.00
     ashamed
    0.99
     asleep
    0.97
     sacked
    0.96
     mad
    0.96
     alright
    0.95
     arrested
    0.94
    Activations Density 0.219%

    No Known Activations