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. 8468
    Prev
    Next
    INDEX
    Explanations

    alphanumeric sequences with decimal numbers

    oai_token-act-pair · gpt-3.5-turbo

    numerical values in various contexts

    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
     vel
    -0.74
     scenery
    -0.63
    icone
    -0.63
     riding
    -0.62
     bleeding
    -0.61
     shopping
    -0.58
     Pros
    -0.57
     travelling
    -0.57
     maiden
    -0.56
     arming
    -0.55
    POSITIVE LOGITS
    134
    0.89
    498
    0.87
    194
    0.86
    132
    0.85
    â̲
    0.85
    708
    0.85
    018
    0.84
    607
    0.83
    148
    0.83
    354
    0.83
    Activations Density 0.213%

    No Known Activations