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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 0-RES-JB
    6. 11078
    Prev
    Next
    INDEX
    Explanations

    mentions of the drug lord Pablo Escobar in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    references to notorious individuals and themes associated with chaos or disorder

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Schwe
    -0.72
     Creep
    -0.70
    igslist
    -0.67
    ãĥł
    -0.66
    earance
    -0.65
    UCT
    -0.65
    ĪĴ
    -0.64
     creeps
    -0.63
     photoc
    -0.62
     propri
    -0.61
    POSITIVE LOGITS
    mad
    3.02
    obar
    2.89
     Madness
    1.29
     madness
    1.17
    Mad
    1.08
    Madison
    0.78
     merry
    0.76
     Mahjong
    0.75
     Madrid
    0.74
    iano
    0.73
    Activations Density 0.023%

    No Known Activations