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. 19665
    Prev
    Next
    INDEX
    Explanations

    phrases related to processes or actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases that refer to the various ways something is done or achieved

    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
    acerb
    -0.78
     drowning
    -0.75
    bda
    -0.71
    bas
    -0.68
    åij
    -0.64
    kaya
    -0.62
    odium
    -0.62
    topic
    -0.62
    tty
    -0.62
    Guard
    -0.61
    POSITIVE LOGITS
    soever
    0.88
     they
    0.78
     humans
    0.77
     individuals
    0.77
     corporations
    0.73
     we
    0.72
     Europeans
    0.69
     historians
    0.68
     observers
    0.67
     workers
    0.66
    Activations Density 0.114%

    No Known Activations