© 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. 0-RES-JB
    11. 23200
    Prev
    Next
    INDEX
    Explanations

    references to the Latin term "cumulative"

    oai_token-act-pair · gpt-3.5-turbo

    references to cumulative concepts or quantities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ryu
    -0.70
    LED
    -0.69
    anka
    -0.69
    tty
    -0.69
    Blade
    -0.67
     Rogue
    -0.66
    Kid
    -0.66
     selector
    -0.65
     Metroid
    -0.65
    Sanders
    -0.65
    POSITIVE LOGITS
     Cum
    4.12
     cum
    2.60
    cum
    2.39
     cumulative
    1.96
     ejac
    1.15
    ulative
    1.07
     Cummings
    1.05
     Dum
    1.05
    aque
    1.00
     Buk
    0.98
    Activations Density 0.040%

    No Known Activations