© 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. 1134
    Prev
    Next
    INDEX
    Explanations

    phrases with the word "way" that represent overcoming challenges or obstacles

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to navigating or moving through situations or challenges

    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
    urses
    -0.69
    hem
    -0.65
    arus
    -0.64
    uzzle
    -0.64
     Mub
    -0.63
     Tuc
    -0.62
     backdrop
    -0.60
     Kau
    -0.60
    Services
    -0.58
    querque
    -0.57
    POSITIVE LOGITS
    ward
    1.14
     into
    1.05
     onto
    1.03
     toward
    1.01
     towards
    1.00
     INTO
    0.95
     forward
    0.90
     ashore
    0.89
     Into
    0.86
    forward
    0.85
    Activations Density 0.047%

    No Known Activations