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. 8458
    Prev
    Next
    INDEX
    Explanations

    specific numerical values or codes, specifically those containing the number 67

    oai_token-act-pair · gpt-3.5-turbo

    numerical identifiers, particularly relating to a specific data set or reference

    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
    ¥ŀ
    -0.91
    achus
    -0.86
    skirts
    -0.85
    mble
    -0.82
    aptic
    -0.81
     unden
    -0.81
     tiss
    -0.78
     tradem
    -0.76
     tension
    -0.71
    lined
    -0.70
    POSITIVE LOGITS
     ILCS
    1.05
    MHz
    0.87
    328
    0.83
    00
    0.81
    67
    0.80
    83
    0.78
    naire
    0.78
    324
    0.77
    344
    0.77
    440
    0.76
    Activations Density 0.023%

    No Known Activations