© 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. 0-RES-JB
    11. 11729
    Prev
    Next
    INDEX
    Explanations

    references to portions of something, whether it be portions of food, text, proceeds, or even the population

    oai_token-act-pair · gpt-3.5-turbo

    references to portions or segments of something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    omics
    -0.68
     Disco
    -0.62
    ante
    -0.61
    urus
    -0.60
    UTC
    -0.59
    thood
    -0.58
    borg
    -0.57
    ittens
    -0.56
     Everton
    -0.56
    grave
    -0.56
    POSITIVE LOGITS
     portion
    3.63
     portions
    2.61
     part
    1.59
     segment
    1.55
     chunk
    1.54
     section
    1.49
     subset
    1.45
    portion
    1.44
     remainder
    1.35
     fraction
    1.31
    Activations Density 0.008%

    No Known Activations