© 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. 2-RES-JB
    11. 2428
    Prev
    Next
    INDEX
    Explanations

    references to numbers and codes within a text

    oai_token-act-pair · gpt-3.5-turbo

    numerical values or data points

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yang
    -0.79
    ezvous
    -0.78
    eering
    -0.74
    BOOK
    -0.70
    furt
    -0.70
    gregation
    -0.67
    yz
    -0.65
    gets
    -0.64
    gments
    -0.64
    feature
    -0.64
    POSITIVE LOGITS
     ILCS
    1.06
    icum
    0.89
    00
    0.88
    80
    0.88
    00000
    0.86
    807
    0.85
    70
    0.84
    319
    0.84
    75
    0.83
    758
    0.79
    Activations Density 0.033%

    No Known Activations