© 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. 16281
    Prev
    Next
    INDEX
    Explanations

    terms related to specific companies and brands, such as "Coca-Cola" and "Mercedes-Benz," often with a negative connotation

    oai_token-act-pair · gpt-3.5-turbo

    references to specific brands and negative contexts associated with them

    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
     prol
    -0.82
     Survivors
    -0.74
     hapl
    -0.72
     study
    -0.71
     barric
    -0.70
     Visitors
    -0.69
     replay
    -0.68
     prophecy
    -0.67
     circum
    -0.67
     migr
    -0.66
    POSITIVE LOGITS
    Cola
    1.59
    Benz
    1.36
    operated
    0.94
    Studio
    0.92
    Col
    0.92
    Columb
    0.90
    Sol
    0.88
    Ch
    0.86
    Bra
    0.86
    Mobil
    0.86
    Activations Density 0.063%

    No Known Activations