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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 10863
    Prev
    Next
    INDEX
    Explanations

    documents that mention the act of destroying or tearing something apart

    oai_token-act-pair · gpt-3.5-turbo

    terms related to fragmentation or tearing apart, particularly in a figurative or literal sense

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Kamp
    -0.69
    xon
    -0.69
     Nou
    -0.68
    stown
    -0.66
     Belg
    -0.64
    ague
    -0.63
    asia
    -0.62
    pect
    -0.61
     Parkinson
    -0.61
    lda
    -0.60
    POSITIVE LOGITS
    ding
    1.38
     shred
    0.99
    sburg
    0.97
    aby
    0.92
     tremend
    0.87
    s
    0.86
    itionally
    0.83
    soever
    0.81
    anguage
    0.81
     horm
    0.80
    Activations Density 0.024%

    No Known Activations