© 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. 10775
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal issues and controversial events involving potential consequences

    oai_token-act-pair · gpt-3.5-turbo

    instances of punctuation, specifically commas

    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
    atorium
    -0.69
    ahu
    -0.66
    wyn
    -0.62
    ICA
    -0.61
    iam
    -0.59
    shore
    -0.56
    chin
    -0.54
    ffen
    -0.53
    USS
    -0.52
    rar
    -0.52
    POSITIVE LOGITS
     namely
    1.31
     albeit
    1.20
     whereas
    1.09
     despite
    1.06
     viz
    1.04
     irrespective
    1.00
     thereby
    0.99
     regardless
    0.96
    albeit
    0.93
     rather
    0.91
    Activations Density 0.407%

    No Known Activations