© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 24565
    Prev
    Next
    INDEX
    Explanations

    names of specific individuals involved in news events

    oai_token-act-pair · gpt-3.5-turbo

    names and proper nouns associated with individuals and specific roles

    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
    acters
    -0.66
     WOM
    -0.65
     robbers
    -0.63
     phantom
    -0.62
     Glac
    -0.61
     sleeper
    -0.61
     oppress
    -0.60
     bandits
    -0.60
     Catalonia
    -0.59
     philosophers
    -0.59
    POSITIVE LOGITS
     Jr
    1.06
     oversaw
    0.99
     oversees
    0.94
    meier
    0.94
     acknowledged
    0.89
     told
    0.88
     presided
    0.87
    iewicz
    0.87
     briefed
    0.87
     said
    0.86
    Activations Density 0.266%

    No Known Activations