© 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. 625
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a contrast or comparison

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Unlike" to highlight comparisons or contrasts

    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
     correction
    -0.67
     supply
    -0.65
     part
    -0.65
     bleeding
    -0.62
     continuing
    -0.60
     line
    -0.59
     talking
    -0.59
     applause
    -0.59
     corrections
    -0.58
     back
    -0.58
    POSITIVE LOGITS
     Unlike
    2.92
    Unlike
    2.43
     Like
    1.63
    Like
    1.59
     Whereas
    1.59
     Compared
    1.46
     Contrary
    1.42
     Notably
    1.42
     Typically
    1.41
     Essentially
    1.40
    Activations Density 0.021%

    No Known Activations