INDEX

Explanations

medical conditions, bodily parts, and substances commonly measured or mentioned in healthcare contexts

oai_token-act-pair · gpt-3.5-turbo

terms related to health, substances, and consumption

oai_token-act-pair · gpt-4o-mini Triggered by @bot

word2 in "word1 and/or word2", mostly for nouns

Explanation Uploaded by User

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

•L9 resid features following 'and'

No Comments

Negative Logits

heit

-0.84

antha

-0.74

iov

-0.73

ariat

-0.73

omaly

-0.71

imum

-0.70

assis

-0.70

undown

-0.70

ËĪ

-0.70

culosis

-0.69

POSITIVE LOGITS

 alike

1.49

 respectively

1.35

 industries

0.92

 striped

0.92

 sectors

0.90

 combo

0.82

 interchange

0.79

 accessories

0.74

 products

0.73

 depending

0.72

Activations Density 0.371%

medical conditions, bodily parts, and substances commonly measured or mentioned in healthcare contexts

terms related to health, substances, and consumption

word2 in "word1 and/or word2", mostly for nouns

No Comments

No Known Activations