© 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. 16469
    Prev
    Next
    INDEX
    Explanations

    phrases emphasizing a point or contrasting expectations

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "not that" followed by a negative qualifier

    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
    姫
    -0.87
    Dialogue
    -0.80
    izont
    -0.74
    Ń·
    -0.73
    rees
    -0.71
    afety
    -0.69
    rahim
    -0.69
    ãĥ¼ãĥ³
    -0.68
    Leaks
    -0.68
    ulus
    -0.68
    POSITIVE LOGITS
     simple
    1.11
     easy
    1.00
     uncommon
    0.92
     kind
    0.91
    simple
    0.90
     way
    0.88
     important
    0.86
     fateful
    0.84
     unusual
    0.83
     complicated
    0.83
    Activations Density 0.059%

    No Known Activations