© 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. 16805
    Prev
    Next
    INDEX
    Explanations

    adjectives related to emotional states, such as 'nurturing' and 'caring'

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the concept of caring or nurturing

    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
    nesota
    -0.72
     prohibition
    -0.70
    ãĥĪ
    -0.67
    quire
    -0.66
    sure
    -0.66
     confirmation
    -0.65
    cluding
    -0.63
    enty
    -0.62
     defense
    -0.62
     cellul
    -0.61
    POSITIVE LOGITS
     cared
    1.10
    lessly
    0.82
    taker
    0.80
    lest
    0.74
    lington
    0.74
     passionately
    0.72
     Adults
    0.72
    amily
    0.71
     citiz
    0.69
     Worker
    0.67
    Activations Density 0.014%

    No Known Activations