© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 24061
    Prev
    Next
    INDEX
    Explanations

    references to mixed martial arts (MMA) and Ultimate Fighting Championship (UFC) events and technology products like iPhones and Android devices

    oai_token-act-pair · gpt-3.5-turbo

    references to mixed martial arts and technology products

    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
    ++++++++++++++++
    -0.62
    âķIJâķIJ
    -0.59
    hap
    -0.58
     âĢº
    -0.56
    ï¸ı
    -0.56
    -------------
    -0.54
    20439
    -0.53
    aws
    -0.51
    beat
    -0.51
    ----------
    -0.51
    POSITIVE LOGITS
     respectively
    0.64
     peripher
    0.62
    ItemThumbnailImage
    0.58
     mosqu
    0.56
    anooga
    0.55
     Citadel
    0.53
     citiz
    0.52
     powerhouse
    0.51
    senal
    0.51
     equivalents
    0.51
    Activations Density 1.457%

    No Known Activations