© 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. 23429
    Prev
    Next
    INDEX
    Explanations

    phrases related to leaving or departing

    oai_token-act-pair · gpt-3.5-turbo

    references to departures or changes in context

    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
    ibur
    -0.77
    essee
    -0.73
    iop
    -0.70
    izable
    -0.69
    illance
    -0.67
    iola
    -0.67
    acca
    -0.66
    paren
    -0.65
    piracy
    -0.64
    icum
    -0.64
    POSITIVE LOGITS
     undone
    0.89
     from
    0.82
    untled
    0.72
    ments
    0.72
    lust
    0.72
    doms
    0.70
    soever
    0.66
     FROM
    0.66
    ï¸
    0.65
    from
    0.64
    Activations Density 0.081%

    No Known Activations