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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 13763
    Prev
    Next
    INDEX
    Explanations

    assertions and allegations along with evidence or lack thereof to support them

    oai_token-act-pair · gpt-3.5-turbo

    phrases that discuss the authenticity or credibility of claims and evidence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     actionGroup
    -0.94
     disadvantage
    -0.67
     downtime
    -0.66
     mobility
    -0.65
     hosp
    -0.65
     specialization
    -0.64
     leisure
    -0.63
     guiActiveUnfocused
    -0.63
     Compact
    -0.62
     restless
    -0.62
    POSITIVE LOGITS
     assertions
    1.34
     claims
    1.19
    Claim
    1.16
     allegations
    1.15
    False
    1.14
     untrue
    1.11
     assertion
    1.10
     debunked
    1.10
     allegation
    1.09
     falsehood
    1.06
    Activations Density 0.340%

    No Known Activations