© 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. 5451
    Prev
    Next
    INDEX
    Explanations

    words related to errors or regrets

    oai_token-act-pair · gpt-3.5-turbo

    phrases that denote errors or mistakes

    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
    rollers
    -0.85
    ighth
    -0.80
    well
    -0.79
    electric
    -0.79
    tsky
    -0.72
    amen
    -0.71
    weeney
    -0.70
    vez
    -0.70
    types
    -0.68
    raq
    -0.68
    POSITIVE LOGITS
     mistake
    1.00
    ously
    0.89
     mistakes
    0.89
     unfocusedRange
    0.82
     mistaken
    0.76
     errors
    0.73
     Malf
    0.71
    lessly
    0.70
     CVE
    0.70
    Maker
    0.69
    Activations Density 0.020%

    No Known Activations