© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 6-RES-JB
    11. 2193
    Prev
    Next
    INDEX
    Explanations

    instances where communication or conversations are happening, specifically with the word 'with' before the entity being interacted with

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "speak with."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fighter
    -0.79
    ©¶æ¥µ
    -0.72
    waters
    -0.71
    ulo
    -0.68
    gross
    -0.65
    wikipedia
    -0.64
    chart
    -0.63
    mol
    -0.63
    anski
    -0.63
    works
    -0.62
    POSITIVE LOGITS
     regards
    1.11
     regard
    1.03
     strangers
    0.89
     passers
    0.85
     Samantha
    0.84
     representatives
    0.84
     coworkers
    0.83
    stood
    0.82
     fellow
    0.80
     constituents
    0.78
    Activations Density 0.102%

    No Known Activations