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

    Neuronpedia

    APICircuit 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. 4-RES-JB
    6. 3370
    Prev
    Next
    INDEX
    Explanations

    legal and political terms related to legal consequences and actions, such as extradition, impeachment, and divorce

    oai_token-act-pair · gpt-3.5-turbo

    terms related to legal actions involving extradition, impeachment, and divorce

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

    No Comments

    Negative Logits
    izen
    -0.84
    olen
    -0.81
    ebus
    -0.80
    acca
    -0.79
    ells
    -0.77
    gm
    -0.76
    mic
    -0.74
    llular
    -0.74
    Cola
    -0.73
    est
    -0.73
    POSITIVE LOGITS
     prevention
    1.08
     proceedings
    0.98
     relief
    0.82
     stemming
    0.80
     initiation
    0.77
     compulsion
    0.76
     Prevention
    0.76
     awards
    0.76
     tribunal
    0.74
     headache
    0.73
    Activations Density 0.044%

    No Known Activations