INDEX

Explanations

words related to the medical field and healthcare, specifically terms related to medical conditions, treatments, and health-related concepts

oai_token-act-pair · gpt-3.5-turbo

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

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+.

-0.72

!.

-0.68

$.

-0.67

 respectively

-0.65

zan

-0.65

().

-0.65

.(

-0.62

.):

-0.61

gettable

-0.60

-0.59

POSITIVE LOGITS

 constitutes

0.97

 depends

0.96

 creates

0.96

 varies

0.96

 ensures

0.96

 undermines

0.95

 enables

0.95

 tends

0.95

 lends

0.93

 reflects

0.93

Activations Density 0.333%

words related to the medical field and healthcare, specifically terms related to medical conditions, treatments, and health-related concepts

No Comments

No Known Activations