© 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. 24335
    Prev
    Next
    INDEX
    Explanations

    phrases related to sending messages or communicating a strong or clear intent

    oai_token-act-pair · gpt-3.5-turbo

    references to messages intended for communication and influence

    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
    ternity
    -0.74
    utsu
    -0.71
     Carbuncle
    -0.70
    ancies
    -0.69
    eger
    -0.67
    ources
    -0.65
    pes
    -0.64
    idia
    -0.62
    illions
    -0.61
    opes
    -0.61
    POSITIVE LOGITS
     loud
    0.97
     message
    0.90
     about
    0.88
     deterrence
    0.83
     louder
    0.83
     loudly
    0.79
     ABOUT
    0.79
    ãģ®éŃĶ
    0.77
    matic
    0.75
     forcefully
    0.74
    Activations Density 0.120%

    No Known Activations