© 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. 10-RES-JB
    11. 18906
    Prev
    Next
    INDEX
    Explanations

    phrases related to certainty and clarity

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and phrases indicating cumulative or non-exclusive conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    swick
    -0.83
    PsyNetMessage
    -0.82
     RPGs
    -0.79
    RPG
    -0.78
     STATS
    -0.70
     Tycoon
    -0.70
    izards
    -0.70
    AGE
    -0.68
     Hou
    -0.67
    luck
    -0.67
    POSITIVE LOGITS
     unconditional
    1.29
     unequiv
    1.22
     palpable
    1.18
     deliberate
    1.15
     decisive
    1.14
     unamb
    1.10
     lasting
    1.10
     immediate
    1.08
     undeniable
    1.07
     conclusive
    1.05
    Activations Density 0.119%

    No Known Activations