© 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
    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. 11394
    Prev
    Next
    INDEX
    Explanations

    phrases related to truth or accuracy

    oai_token-act-pair · gpt-3.5-turbo

    statements about truth or verification

    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
    ercise
    -0.80
    isphere
    -0.75
    igan
    -0.74
    reau
    -0.66
    cloth
    -0.65
     Habit
    -0.63
    wal
    -0.62
    boy
    -0.58
    aban
    -0.58
    lease
    -0.56
    POSITIVE LOGITS
     true
    3.66
    true
    2.82
    True
    2.22
     TRUE
    2.15
     True
    2.11
     false
    1.85
    false
    1.77
     truth
    1.60
     untrue
    1.47
    False
    1.44
    Activations Density 0.027%

    No Known Activations