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. 4-RES-JB
    6. 8118
    Prev
    Next
    INDEX
    Explanations

    names of government agencies and organizations

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns related to organizations, entities, and notable subjects

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

    No Comments

    Negative Logits
    Enlarge
    -0.59
    î
    -0.53
     ..."
    -0.50
    ``
    -0.50
    mx
    -0.47
    chel
    -0.47
    la
    -0.47
    berra
    -0.46
    upper
    -0.46
    âķIJâķIJ
    -0.46
    POSITIVE LOGITS
     cannot
    0.65
     reverted
    0.64
     underwent
    0.63
     succeeded
    0.63
     countered
    0.61
     rul
    0.60
     experimented
    0.56
     redeemed
    0.56
     ceased
    0.55
     shifted
    0.55
    Activations Density 1.043%

    No Known Activations