© 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. 3-RES-JB
    11. 20793
    Prev
    Next
    INDEX
    Explanations

    words related to criticism or disapproval

    oai_token-act-pair · gpt-3.5-turbo

    terms related to expressing disapproval or condemnation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ramid
    -0.90
     Solitaire
    -0.78
    thora
    -0.73
    iago
    -0.72
    opic
    -0.71
    vati
    -0.70
    ovember
    -0.70
    weeney
    -0.69
    aldo
    -0.69
    hack
    -0.67
    POSITIVE LOGITS
     condemn
    0.98
     condemning
    0.93
     condemns
    0.85
     condemnation
    0.85
    urous
    0.84
     condemned
    0.78
    代
    0.74
     denounce
    0.74
    uses
    0.72
    terday
    0.71
    Activations Density 0.014%

    No Known Activations