© 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. 11-RES-JB
    11. 3308
    Prev
    Next
    INDEX
    Explanations

    action verbs indicating performance or behavior

    oai_token-act-pair · gpt-3.5-turbo

    phrases that reference actions or behaviors being discussed

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    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
    usk
    -0.72
    isphere
    -0.65
    addon
    -0.64
    selection
    -0.61
    inen
    -0.61
    atto
    -0.60
    iHUD
    -0.60
    irl
    -0.59
    osa
    -0.58
    apsed
    -0.58
    POSITIVE LOGITS
     deserve
    1.13
     offend
    1.07
     mattered
    1.02
     aren
    1.00
     resonate
    0.95
     weren
    0.93
     nobody
    0.93
     enrich
    0.92
    rouse
    0.91
     shouldn
    0.90
    Activations Density 0.188%

    No Known Activations