© 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 EvalsExportsinterp-engineNEWAPI 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. 12-RES-JB
    11. 3606
    Prev
    Next
    INDEX
    Explanations

    names of people or characters in a narrative, specifically with mentions of their emotions or actions

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    instances of emotional distress in legal or tragic situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãģ®éŃĶ
    -0.76
    igmatic
    -0.74
    bard
    -0.73
    vantage
    -0.70
    +)
    -0.70
    soever
    -0.69
    ourage
    -0.68
    ictionary
    -0.66
    pection
    -0.66
    ï¸ı
    -0.66
    POSITIVE LOGITS
     how
    1.21
     seeing
    1.16
     losing
    1.02
     witnessing
    1.01
     having
    1.00
     hearing
    0.99
     what
    0.96
     being
    0.93
     receiving
    0.92
     encountering
    0.86
    Activations Density 0.254%

    No Known Activations