© 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. 4650
    Prev
    Next
    INDEX
    Explanations

    comparative adjectives

    oai_token-act-pair · gpt-3.5-turbo

    descriptive adjectives that indicate improvement or enhancement

    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
    este
    -0.80
    odor
    -0.75
    mad
    -0.75
    ainted
    -0.74
    etition
    -0.72
    EP
    -0.69
    tra
    -0.69
    Christmas
    -0.68
    EVA
    -0.68
    eno
    -0.67
    POSITIVE LOGITS
     than
    1.62
     Than
    1.43
    than
    1.22
     "$:/
    1.00
     ado
    0.84
     millenn
    0.80
     sibling
    0.77
    ModLoader
    0.76
     Seym
    0.75
    getic
    0.75
    Activations Density 0.039%

    No Known Activations