© 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. 4-RES-JB
    11. 14716
    Prev
    Next
    INDEX
    Explanations

    phrases related to things that are seen as not needed or avoidable

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating redundancy or excess

    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
    ebus
    -0.91
    deck
    -0.83
    eta
    -0.83
    ership
    -0.80
    ning
    -0.79
    ois
    -0.76
    ramid
    -0.76
    estamp
    -0.74
    imore
    -0.74
    robe
    -0.72
    POSITIVE LOGITS
     unnecessary
    1.11
     unnecess
    0.98
     duplication
    0.93
     precaution
    0.89
     waste
    0.87
     complication
    0.87
     expense
    0.87
     superflu
    0.85
     burdens
    0.82
     burden
    0.81
    Activations Density 0.017%

    No Known Activations