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. 12-RES-JB
    6. 14514
    Prev
    Next
    INDEX
    Explanations

    words and phrases related to criminal activity and punishment

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    words related to criminal acts or offenses

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    gypt
    -0.70
    etsk
    -0.70
    DCS
    -0.69
    ãĤ¼ãĤ¦ãĤ¹
    -0.69
    framework
    -0.66
     Appl
    -0.66
     partName
    -0.64
    itsch
    -0.64
    ::
    -0.64
     Mehran
    -0.63
    POSITIVE LOGITS
     suicide
    1.44
     perjury
    1.35
     adultery
    1.35
     crimes
    1.30
     heinous
    1.22
     atrocities
    1.20
     treason
    1.13
     misdem
    1.13
     genocide
    1.12
     manslaughter
    1.12
    Activations Density 0.082%

    No Known Activations