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. 3-RES-JB
    6. 19968
    Prev
    Next
    INDEX
    Explanations

    mentions of proper nouns related to people or places, particularly focusing on the word "Newton"

    oai_token-act-pair · gpt-3.5-turbo

    references to the name "Newton"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    inion
    -0.78
    RAW
    -0.73
    ħĭ
    -0.71
    iov
    -0.71
    pport
    -0.71
    orescent
    -0.68
    reement
    -0.68
    Veter
    -0.68
    Hispanic
    -0.67
    olulu
    -0.66
    POSITIVE LOGITS
     Newton
    1.15
    ©¶æ¥µ
    0.89
    ufact
    0.74
    sburg
    0.73
     Ragnarok
    0.73
     Keynes
    0.71
     Medal
    0.71
    enberg
    0.69
     Collider
    0.67
    batch
    0.66
    Activations Density 0.004%

    No Known Activations