© 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
    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. 1204
    Prev
    Next
    INDEX
    Explanations

    instances of the word "finding."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the word "finding."

    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
    ttle
    -0.68
    ened
    -0.62
    enary
    -0.60
     Administ
    -0.60
    rior
    -0.59
    ep
    -0.59
    erves
    -0.57
    Ill
    -0.56
    plate
    -0.55
    nas
    -0.55
    POSITIVE LOGITS
     finding
    3.57
     discovering
    2.24
    finding
    2.19
     Finding
    2.16
    Finding
    2.13
     locating
    1.71
     figuring
    1.70
    find
    1.57
     find
    1.51
     findings
    1.48
    Activations Density 0.016%

    No Known Activations