© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 10094
    Prev
    Next
    INDEX
    Explanations

    phrases related to controversial or negative statements/actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to legal proceedings and allegations of misconduct

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    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
    issance
    -0.90
    FIG
    -0.85
    erenn
    -0.78
    ugu
    -0.78
    hov
    -0.75
    âĻ
    -0.75
    imil
    -0.74
    HD
    -0.74
    Ñĭ
    -0.72
    Hist
    -0.72
    POSITIVE LOGITS
     relation
    1.22
     regards
    1.14
     awarding
    1.13
     dealings
    1.11
     favor
    1.03
     regard
    1.02
     connection
    1.01
     hiring
    0.98
     interviews
    0.98
     lieu
    0.98
    Activations Density 0.238%

    No Known Activations