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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 9-RES-JB
    6. 22361
    Prev
    Next
    INDEX
    Explanations

    phrases related to being alive, active or operational, and phrases related to reopening and keeping something in operation

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions that link ideas or phrases about being alive or operational status

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æĥ
    -0.75
    udence
    -0.69
    conserv
    -0.69
    UE
    -0.69
    urities
    -0.68
     ç¥ŀ
    -0.68
    anmar
    -0.68
    Shape
    -0.67
    AMA
    -0.66
    STEM
    -0.66
    POSITIVE LOGITS
     ready
    1.50
     awaiting
    1.24
     waiting
    1.10
     poised
    1.09
     willing
    1.05
     alive
    1.02
     itching
    1.02
     primed
    1.00
     eager
    1.00
     hungry
    1.00
    Activations Density 0.252%

    No Known Activations