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. 3-RES-JB
    6. 22189
    Prev
    Next
    INDEX
    Explanations

    words related to large amounts of movement or intake

    oai_token-act-pair · gpt-3.5-turbo

    terms related to large movements or increases in groups, particularly in the context of people or goods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ovember
    -0.78
    MET
    -0.74
    aple
    -0.73
    fman
    -0.70
    acter
    -0.67
    Ingredients
    -0.66
    izons
    -0.65
     Atomic
    -0.65
    Cola
    -0.64
    zan
    -0.64
    POSITIVE LOGITS
     influx
    0.99
     flooding
    0.88
    eson
    0.85
    uating
    0.84
     arrivals
    0.81
    naires
    0.80
     infl
    0.80
     flows
    0.78
    ãĥ¼ãĥĨãĤ£
    0.77
    uates
    0.76
    Activations Density 0.014%

    No Known Activations