© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 8097
    Prev
    Next
    INDEX
    Explanations

    superlatives, specifically the word "most" followed by a number indicating a high degree of comparison

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the concept of 'most' or 'highest' in comparative contexts

    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
     committed
    -0.66
     opportunity
    -0.60
     signed
    -0.58
    uzz
    -0.57
     Day
    -0.55
     guaranteed
    -0.54
     different
    -0.53
     II
    -0.53
     impressed
    -0.53
     tart
    -0.52
    POSITIVE LOGITS
    most
    4.40
    highest
    1.78
    liest
    1.49
    largest
    1.44
    best
    1.37
    worst
    1.31
    Most
    1.25
    iest
    1.19
    almost
    1.14
    few
    1.13
    Activations Density 0.009%

    No Known Activations