© 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. 12131
    Prev
    Next
    INDEX
    Explanations

    phrases indicating responsibility or decision-making authority

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "up" indicating intensity or physical position

    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
     Blues
    -0.70
     CLSID
    -0.69
    gemony
    -0.64
    gee
    -0.64
    "]=>
    -0.63
     !=
    -0.63
    9999
    -0.60
     Abyss
    -0.59
    [/
    -0.58
    >]
    -0.58
    POSITIVE LOGITS
    stairs
    0.91
    river
    0.85
    dating
    0.84
    grading
    0.81
    graded
    0.80
    rated
    0.79
    skirts
    0.77
    ended
    0.76
    law
    0.75
    acebook
    0.74
    Activations Density 0.055%

    No Known Activations