© 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. 22582
    Prev
    Next
    INDEX
    Explanations

    news-related terms or phrases

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating video watching or viewing experiences

    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
    eatures
    -0.90
    vertisements
    -0.88
     endeav
    -0.88
     arrang
    -0.87
     explan
    -0.84
    aciously
    -0.84
     facult
    -0.83
     psychiat
    -0.82
    fortunately
    -0.81
     behav
    -0.80
    POSITIVE LOGITS
     Latest
    1.26
     Timeline
    1.14
     Thousands
    1.11
     Could
    1.09
     Report
    1.09
     Hundreds
    1.09
     Highlights
    1.08
     Why
    1.08
     Former
    1.08
     Protesters
    1.07
    Activations Density 0.111%

    No Known Activations