© 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. 1993
    Prev
    Next
    INDEX
    Explanations

    mentions on news channels, specifically MSNBC, Fox News, ABC, CBS, CNN, and NBC

    oai_token-act-pair · gpt-3.5-turbo

    mentions of news networks and their related discussions

    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
    objects
    -0.65
    alties
    -0.64
     envelop
    -0.62
     solitary
    -0.61
    urga
    -0.61
    AGES
    -0.59
    ources
    -0.59
     curs
    -0.58
     STATS
    -0.57
    pires
    -0.57
    POSITIVE LOGITS
     aired
    1.03
     Tonight
    0.99
    ornings
    0.95
     host
    0.94
     Hannity
    0.93
     Morning
    0.89
     interviewer
    0.87
     airing
    0.87
     tonight
    0.86
     MSNBC
    0.85
    Activations Density 0.136%

    No Known Activations