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. 15833
    Prev
    Next
    INDEX
    Explanations

    adjectives describing the physical characteristics of objects

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating quality or comparison about 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
    orously
    -0.69
    letters
    -0.64
    claimed
    -0.62
     Memor
    -0.62
    irs
    -0.60
     Inqu
    -0.60
     vacancies
    -0.58
    gart
    -0.58
     Journals
    -0.57
    OUP
    -0.56
    POSITIVE LOGITS
     but
    0.85
     because
    0.78
    .
    0.76
     BUT
    0.76
     compared
    0.73
     considering
    0.71
     due
    0.70
    but
    0.69
     anyways
    0.68
     :)
    0.68
    Activations Density 0.449%

    No Known Activations