Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 23606
    Prev
    Next
    INDEX
    Explanations

    the word "da" in various contexts or phrases

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific names or keywords associated with events or titles

    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
    Not in Any Lists

    No Comments

    Negative Logits
     liqu
    -0.65
     redevelop
    -0.63
     distilled
    -0.62
     warranted
    -0.61
     neut
    -0.59
     augmented
    -0.59
     range
    -0.58
     penet
    -0.58
     SEAL
    -0.58
     magn
    -0.58
    POSITIVE LOGITS
    da
    4.49
    dal
    1.60
    DA
    1.54
    ds
    1.46
    din
    1.43
    de
    1.43
    Da
    1.38
    d
    1.35
    dan
    1.34
    di
    1.34
    Activations Density 0.012%

    No Known Activations