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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 7439
    Prev
    Next
    INDEX
    Explanations

    the word "Apart" at the start of a sentence

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "apart from" in various contexts

    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
     die
    -0.68
     cascade
    -0.62
     lifetime
    -0.62
     signature
    -0.59
     NOAA
    -0.59
    é¾į
    -0.58
     signaling
    -0.58
     trick
    -0.57
    ability
    -0.57
    ryn
    -0.56
    POSITIVE LOGITS
     Apart
    4.11
    Apart
    3.10
     apart
    1.46
     Aside
    1.46
     apartheid
    1.40
    Aside
    1.22
     Away
    1.04
     Needless
    0.96
     aside
    0.95
     Besides
    0.92
    Activations Density 0.018%

    No Known Activations