© 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. 15598
    Prev
    Next
    INDEX
    Explanations

    mentions of specific names, possibly focusing on a particular person named "Lie" or "Diane."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific individuals, particularly those associated with the name "Lie" and the concept of "idol."

    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
    orsi
    -0.69
    ells
    -0.67
    pmwiki
    -0.67
    eele
    -0.66
    fully
    -0.66
    iates
    -0.66
    icum
    -0.65
    heartedly
    -0.64
    iated
    -0.63
    cially
    -0.62
    POSITIVE LOGITS
    berman
    0.95
     theoret
    0.72
    rection
    0.71
    gu
    0.69
    utenant
    0.69
    ga
    0.69
    braska
    0.68
    vu
    0.68
    gie
    0.67
    glers
    0.65
    Activations Density 0.104%

    No Known Activations