© 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. 0-RES-JB
    11. 23341
    Prev
    Next
    INDEX
    Explanations

    references to the messaging platform "WhatsApp"

    oai_token-act-pair · gpt-3.5-turbo

    references to messaging applications and platforms

    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
     rig
    -0.74
     ven
    -0.72
    Robert
    -0.68
     mineral
    -0.67
     stud
    -0.67
     Robert
    -0.66
    rig
    -0.65
     anchored
    -0.64
    tie
    -0.64
     Henry
    -0.62
    POSITIVE LOGITS
     WhatsApp
    4.00
     Telegram
    2.24
     Whats
    1.99
     Snapchat
    1.98
     Messenger
    1.87
     SMS
    1.73
     Tinder
    1.64
     Skype
    1.57
     Slack
    1.46
     Gmail
    1.46
    Activations Density 0.035%

    No Known Activations