© 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. 8947
    Prev
    Next
    INDEX
    Explanations

    statistical information including sample sizes and margins of error

    oai_token-act-pair · gpt-3.5-turbo

    numeric statistical data related to margins of error and percentages

    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
    dos
    -0.73
     Trance
    -0.72
    ggies
    -0.71
    module
    -0.70
    Reloaded
    -0.69
    chel
    -0.66
    itiz
    -0.66
    warts
    -0.66
    tops
    -0.64
    drivers
    -0.63
    POSITIVE LOGITS
     +/-
    0.99
     margin
    0.77
     RAND
    0.74
     error
    0.73
     methodological
    0.73
     weighted
    0.72
     statistical
    0.72
     margins
    0.71
     coefficient
    0.70
     Correction
    0.68
    Activations Density 0.111%

    No Known Activations