© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 2-RES-JB
    11. 18527
    Prev
    Next
    INDEX
    Explanations

    proper nouns or specific terms in uppercase letters

    oai_token-act-pair · gpt-3.5-turbo

    terms associated with a specific brand or product, as well as concepts of dismantling or restructuring

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nar
    -1.00
    sch
    -0.96
    ends
    -0.90
    sburgh
    -0.86
    enegger
    -0.85
    entin
    -0.85
    nir
    -0.83
    sg
    -0.80
     Images
    -0.79
    s
    -0.79
    POSITIVE LOGITS
    chnology
    0.91
     Polo
    0.76
    BILITY
    0.74
    veyard
    0.72
    abwe
    0.72
     millenn
    0.70
    ãĤ§
    0.70
    breaker
    0.69
     skating
    0.68
    ffiti
    0.68
    Activations Density 0.033%

    No Known Activations