Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 17974
    Prev
    Next
    INDEX
    Explanations

    weights or measurements in kilograms

    oai_token-act-pair · gpt-3.5-turbo

    references to measurements of weight in kilograms

    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
    gotten
    -0.81
     Dee
    -0.62
    irie
    -0.62
    zig
    -0.60
    structed
    -0.60
     Maker
    -0.60
     Meier
    -0.60
     Explain
    -0.60
    bidden
    -0.59
     Fault
    -0.59
    POSITIVE LOGITS
    atsu
    0.86
     gorilla
    0.81
    kg
    0.80
    omez
    0.79
     ammonia
    0.76
    rieg
    0.75
     kg
    0.75
    ross
    0.73
    emouth
    0.73
    eter
    0.73
    Activations Density 0.010%

    No Known Activations