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. 16042
    Prev
    Next
    INDEX
    Explanations

    the word "many" being used across various contexts

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "many" and its variations, indicating a focus on collective opinion or consensus

    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
    ibal
    -0.73
    ĸļ
    -0.72
    rompt
    -0.69
    llor
    -0.69
    ataka
    -0.68
    ifice
    -0.67
    acion
    -0.67
    ocol
    -0.66
    acus
    -0.66
    opy
    -0.65
    POSITIVE LOGITS
     observers
    1.08
     people
    0.97
     onlook
    0.97
     commentators
    0.96
     consumers
    0.93
     Americans
    0.93
     peoples
    0.93
     individuals
    0.92
     households
    0.90
     economists
    0.88
    Activations Density 0.163%

    No Known Activations