© 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. 8-RES-JB
    11. 3331
    Prev
    Next
    INDEX
    Explanations

    technical details or specifications

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "technically" and related variations 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
    ieties
    -0.76
    umbn
    -0.70
    itu
    -0.68
    ĸļ
    -0.67
    ivo
    -0.66
    awaru
    -0.65
     Vaj
    -0.64
    oday
    -0.63
    ravings
    -0.63
    lain
    -0.62
    POSITIVE LOGITS
     correct
    0.86
     speaking
    0.84
     prohibited
    0.81
     incorrect
    0.79
     obligated
    0.79
     minded
    0.76
     inclined
    0.73
     forbidden
    0.73
     incapable
    0.72
     outlawed
    0.70
    Activations Density 0.048%

    No Known Activations