© 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. 4396
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a sequence of events or steps

    oai_token-act-pair · gpt-3.5-turbo

    punctuation, specifically commas

    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
    Cause
    -0.68
    ogl
    -0.66
    Detailed
    -0.64
    iously
    -0.64
    Morning
    -0.63
    ãĥ¥
    -0.63
    Track
    -0.62
    RL
    -0.62
    phony
    -0.60
    Laughs
    -0.60
    POSITIVE LOGITS
     however
    1.16
     though
    0.88
     moreover
    0.83
     albeit
    0.77
     although
    0.72
     alas
    0.71
     please
    0.70
     namely
    0.70
     suffice
    0.67
     barring
    0.66
    Activations Density 0.405%

    No Known Activations