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. 4-RES-JB
    6. 18192
    Prev
    Next
    INDEX
    Explanations

    phrases related to communication or conveying information

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "communicate" and its variations

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

    No Comments

    Negative Logits
    uck
    -0.83
    steen
    -0.80
    Fan
    -0.76
    fall
    -0.76
    gger
    -0.73
    sm
    -0.73
    este
    -0.73
    strap
    -0.72
    Lago
    -0.72
    liga
    -0.71
    POSITIVE LOGITS
     behavi
    0.93
     communicates
    0.92
     communication
    0.90
     communicated
    0.84
     communicating
    0.79
     Communication
    0.78
     warr
    0.78
    iments
    0.77
     messages
    0.77
     communicate
    0.74
    Activations Density 0.015%

    No Known Activations