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. 11778
    Prev
    Next
    INDEX
    Explanations

    references to European Space Agency (ESA) or mentions of baking related terms, particularly pastry

    oai_token-act-pair · gpt-3.5-turbo

    references to the European Space Agency (ESA) and pastries

    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
    aroo
    -0.95
    anas
    -0.93
    aho
    -0.92
    ias
    -0.87
    lining
    -0.85
    odes
    -0.84
    oos
    -0.82
    itar
    -0.81
    stone
    -0.80
    gins
    -0.79
    POSITIVE LOGITS
    âĺħâĺħ
    0.78
    ¬¼
    0.78
    chnology
    0.76
    terday
    0.69
    oire
    0.66
    »Ĵ
    0.65
    ++++
    0.65
     Steelers
    0.63
    ++++++++++++++++
    0.62
    BF
    0.62
    Activations Density 0.030%

    No Known Activations