© 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. 18381
    Prev
    Next
    INDEX
    Explanations

    words related to suppression or exerting control

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "suppress" and its derivatives in various contexts of restriction or control

    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
    verty
    -0.66
    member
    -0.66
    Stars
    -0.65
    aldo
    -0.65
    olf
    -0.65
    sky
    -0.63
    joining
    -0.63
    gain
    -0.63
    achel
    -0.63
    ̶
    -0.63
    POSITIVE LOGITS
    ively
    0.98
     suppression
    0.94
     suppress
    0.91
     suppressing
    0.89
     suppressed
    0.81
     inhib
    0.79
     impulses
    0.75
     muzzle
    0.73
     distractions
    0.70
     inhibition
    0.70
    Activations Density 0.024%

    No Known Activations