© 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. 10-RES-JB
    11. 17409
    Prev
    Next
    INDEX
    Explanations

    phrases indicating communication or reporting between individuals

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "told" and related phrases indicating communication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    adesh
    -0.94
     sidx
    -0.73
    ateral
    -0.72
    served
    -0.72
    é¾įåĸļ士
    -0.72
    negie
    -0.71
    conservancy
    -0.70
    avascript
    -0.69
    icons
    -0.69
    arthy
    -0.68
    POSITIVE LOGITS
     him
    1.26
     me
    1.11
     us
    1.11
     them
    1.08
     reporters
    0.94
     jokes
    0.87
     everyone
    0.86
     investigators
    0.86
     superiors
    0.84
     stories
    0.84
    Activations Density 0.072%

    No Known Activations