© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 5205
    Prev
    Next
    INDEX
    Explanations

    claims or mentions related to the concept of "true" alongside proper nouns like names or places

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "True" as it relates to fact-checking or veracity claims

    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
    isphere
    -0.80
     lag
    -0.75
     adjourn
    -0.72
     annex
    -0.66
    pee
    -0.66
    ariat
    -0.63
     econom
    -0.62
    dep
    -0.62
    abi
    -0.62
    aned
    -0.61
    POSITIVE LOGITS
     True
    3.69
    True
    3.38
     False
    2.53
    False
    2.33
     TRUE
    2.10
    true
    1.79
     true
    1.76
     FALSE
    1.39
    false
    1.27
     Truth
    1.21
    Activations Density 0.024%

    No Known Activations