© 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. 4-RES-JB
    11. 22440
    Prev
    Next
    INDEX
    Explanations

    references related to directing actions towards the reader, often with a focus on protection or support

    oai_token-act-pair · gpt-3.5-turbo

    references to the pronoun "you."

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

    No Comments

    Negative Logits
    âĢ¢âĢ¢
    -0.68
     Commerce
    -0.65
    ħ
    -0.65
    amar
    -0.63
    ftime
    -0.62
    asia
    -0.62
    english
    -0.62
    grave
    -0.61
     Cook
    -0.61
    ipal
    -0.61
    POSITIVE LOGITS
     guys
    1.17
    're
    1.06
    tub
    1.04
    hei
    0.91
    RS
    0.90
     know
    0.82
    've
    0.81
    'll
    0.79
     yourselves
    0.73
     yourself
    0.71
    Activations Density 0.151%

    No Known Activations