Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 10323
    Prev
    Next
    INDEX
    Explanations

    the verb "to" at the beginning of sentences, potentially indicating actions or intentions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "to" followed by varying actions or conditions

    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
    starting
    -0.77
    started
    -0.76
    EStream
    -0.74
    Secondly
    -0.69
    quickShipAvailable
    -0.67
     Starts
    -0.67
     starting
    -0.66
     Ending
    -0.66
     endings
    -0.65
     goodbye
    -0.64
    POSITIVE LOGITS
     refine
    1.19
     defy
    1.07
     operate
    1.05
     evolve
    1.05
     struggle
    1.05
     wrestle
    1.04
     insist
    1.03
     pursue
    1.03
     strive
    1.00
     innovate
    0.99
    Activations Density 0.065%

    No Known Activations