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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 5137
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences indicating levels of satisfaction or excitement

    oai_token-act-pair · gpt-3.5-turbo

    negative constructions or expressions of dissatisfaction

    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
     Charity
    -0.56
     ank
    -0.56
     Globe
    -0.56
     leash
    -0.53
    "]=>
    -0.52
     recept
    -0.51
     escal
    -0.50
     subpoena
    -0.50
     acqu
    -0.50
     haunt
    -0.50
    POSITIVE LOGITS
     simpler
    0.93
     clearer
    0.92
     nicer
    0.85
     farther
    0.81
     happier
    0.81
     easier
    0.80
    better
    0.78
     cooler
    0.77
     worse
    0.74
    ł
    0.74
    Activations Density 0.127%

    No Known Activations