© 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. 3-RES-JB
    11. 22633
    Prev
    Next
    INDEX
    Explanations

    mentions or references to the city "Medellin."

    oai_token-act-pair · gpt-3.5-turbo

    references to medical or healthcare-related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ģĸ
    -0.81
    cipline
    -0.79
     innocence
    -0.77
    chnology
    -0.76
    AUT
    -0.73
    âĻ¦
    -0.69
    !/
    -0.68
    ãģį
    -0.68
    RECT
    -0.68
    ESH
    -0.67
    POSITIVE LOGITS
    ications
    1.11
    iator
    1.05
    iterranean
    1.04
    allion
    1.03
    ient
    0.99
    itations
    0.98
    iation
    0.97
    itate
    0.96
    undai
    0.92
    icum
    0.91
    Activations Density 0.009%

    No Known Activations