© 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 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. 3-RES-JB
    11. 20163
    Prev
    Next
    INDEX
    Explanations

    words related to feelings of worry or distress

    oai_token-act-pair · gpt-3.5-turbo

    expressions of anxiety and emotional distress

    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
    OH
    -0.70
    way
    -0.64
    hook
    -0.64
    owan
    -0.63
    omaly
    -0.61
     constituted
    -0.60
    Īè
    -0.60
    wark
    -0.60
    yssey
    -0.59
    Redditor
    -0.59
    POSITIVE LOGITS
     fret
    1.11
    lessly
    0.92
    fully
    0.82
    ful
    0.74
    lect
    0.74
    nir
    0.72
    vier
    0.71
    eking
    0.71
    FUL
    0.71
    lessness
    0.71
    Activations Density 0.018%

    No Known Activations