© 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. 2323
    Prev
    Next
    INDEX
    Explanations

    phrases indicating comparison or showing contrast

    oai_token-act-pair · gpt-3.5-turbo

    the word "few" and its variations indicating scarcity or a small number

    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
    reon
    -0.77
    agher
    -0.70
    dress
    -0.67
    irst
    -0.66
    roup
    -0.65
    ican
    -0.65
    alam
    -0.63
    igmatic
    -0.62
    raine
    -0.61
     Cutter
    -0.61
    POSITIVE LOGITS
     hundred
    0.99
     dozen
    0.92
    een
    0.87
     thousand
    0.82
    uously
    0.82
     mortals
    0.79
     milliseconds
    0.76
    est
    0.76
    ishers
    0.74
    eteen
    0.72
    Activations Density 0.027%

    No Known Activations