© 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. 22636
    Prev
    Next
    INDEX
    Explanations

    phrases related to formal statements or declarations

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "said in," indicating references to statements or quotes

    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
    76561
    -0.84
     %%
    -0.63
    killed
    -0.61
    $$$$
    -0.61
     needs
    -0.61
     needles
    -0.61
     manip
    -0.60
     snipers
    -0.60
     egreg
    -0.60
     ens
    -0.59
    POSITIVE LOGITS
     unison
    1.13
     lieu
    1.05
    clus
    1.01
     conjunction
    0.98
     response
    0.95
    aug
    0.91
    cluded
    0.91
     advance
    0.91
    humane
    0.90
     accordance
    0.89
    Activations Density 0.087%

    No Known Activations