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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & 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. 8-RES-JB
    6. 18897
    Prev
    Next
    INDEX
    Explanations

    verbs related to being discovered or apprehended

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "caught."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ented
    -0.64
     delim
    -0.63
     sclerosis
    -0.63
    psey
    -0.62
    ankind
    -0.59
     ceremonies
    -0.59
    aceutical
    -0.58
    estate
    -0.58
     atmosp
    -0.58
     columns
    -0.58
    POSITIVE LOGITS
     unprepared
    0.88
     plagiar
    0.79
    ãĤ¤ãĥĪ
    0.72
     tampering
    0.67
    netflix
    0.67
    ãĥ³ãĤ¸
    0.67
    coni
    0.66
    bryce
    0.64
    bys
    0.64
     dere
    0.64
    Activations Density 0.026%

    No Known Activations