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. 8-RES-JB
    6. 3500
    Prev
    Next
    INDEX
    Explanations

    phrases discussing past events or experiences

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include "what happened to" followed by various subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     guiActiveUn
    -0.76
    voy
    -0.66
     parting
    -0.60
    pants
    -0.58
    hall
    -0.58
    going
    -0.58
     Uses
    -0.58
    soDeliveryDate
    -0.58
     consulted
    -0.58
    urnal
    -0.56
    POSITIVE LOGITS
    othy
    0.89
     deserve
    0.77
     compensate
    0.76
    pless
    0.75
     regain
    0.74
     stabilize
    0.73
     commemorate
    0.73
     them
    0.72
     accumulate
    0.69
     resemble
    0.69
    Activations Density 0.089%

    No Known Activations