© 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. 8-RES-JB
    11. 879
    Prev
    Next
    INDEX
    Explanations

    verb phrases related to communication or interaction

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "hasn't spoken" or similar expressions indicating a lack of communication

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nen
    -0.66
    Eastern
    -0.62
    inki
    -0.59
    Protect
    -0.58
    accompanied
    -0.58
    prone
    -0.56
    nee
    -0.55
    den
    -0.54
    nick
    -0.54
    adj
    -0.54
    POSITIVE LOGITS
     nor
    1.25
     yet
    1.10
    yet
    1.08
     anymore
    1.00
     since
    0.99
     anything
    0.95
     anybody
    0.95
     anywhere
    0.89
     anyone
    0.86
     sufficiently
    0.85
    Activations Density 0.313%

    No Known Activations