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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 23811
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a desire or action to push boundaries or take actions to a greater extent

    oai_token-act-pair · gpt-3.5-turbo

    expressions indicating progression or an increase in intensity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    pected
    -0.74
    other
    -0.69
     Cust
    -0.65
    ech
    -0.62
    esi
    -0.61
    ogy
    -0.59
     EMP
    -0.59
     Warcraft
    -0.59
     Sorce
    -0.58
    SEA
    -0.58
    POSITIVE LOGITS
     than
    1.05
     distances
    0.90
    than
    0.82
    fetched
    0.77
    backer
    0.73
     awa
    0.73
    ////////////////
    0.69
     towards
    0.69
     distance
    0.68
     overboard
    0.66
    Activations Density 0.055%

    No Known Activations