© 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. 8-RES-JB
    11. 20100
    Prev
    Next
    INDEX
    Explanations

    phrases related to syndicated shows or content

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "syndicate" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    naissance
    -0.77
    ception
    -0.77
    horn
    -0.71
    âĨij
    -0.68
    kus
    -0.68
     Advocate
    -0.68
    jee
    -0.66
    ++++++++++++++++
    -0.63
     Malt
    -0.62
    ksh
    -0.61
    POSITIVE LOGITS
    icate
    1.00
    odic
    0.90
    ocrat
    0.89
    archy
    0.88
    reddits
    0.87
    imore
    0.87
    anners
    0.86
    ony
    0.85
    owship
    0.83
    asma
    0.80
    Activations Density 0.094%

    No Known Activations