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. 0-RES-JB
    6. 24438
    Prev
    Next
    INDEX
    Explanations

    occurrences of the number 67

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the number 67 in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •Embedding Space Digit Features

    No Comments

    Negative Logits
    Newsletter
    -0.81
    tumblr
    -0.68
    taboola
    -0.68
    SpaceEngineers
    -0.68
    éĽ
    -0.67
    soDeliveryDate
    -0.66
    udence
    -0.65
    æĢ
    -0.65
    Tesla
    -0.64
    dyl
    -0.64
    POSITIVE LOGITS
     67
    2.60
     68
    2.58
     73
    2.55
     71
    2.55
     66
    2.55
     77
    2.54
     69
    2.54
     74
    2.52
     76
    2.50
     79
    2.44
    Activations Density 0.060%

    No Known Activations