© 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. 10726
    Prev
    Next
    INDEX
    Explanations

    words or symbols associated with an ending or a conclusion

    oai_token-act-pair · gpt-3.5-turbo

    instances of a specific symbol or character sequence, likely indicating a formatting error or unusual character input

    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
     obser
    -0.81
     puff
    -0.80
     imperson
    -0.75
     ende
    -0.72
     imagination
    -0.71
     unconscious
    -0.71
     endings
    -0.68
     dossier
    -0.68
     idea
    -0.67
     circulation
    -0.66
    POSITIVE LOGITS
    ¯
    1.01
    ï¸ı
    1.00
    âĢł
    0.90
    said
    0.88
    °
    0.86
    âĻ¦
    0.85
    east
    0.84
    âĪ
    0.82
    £
    0.80
    STEM
    0.79
    Activations Density 0.248%

    No Known Activations