INDEX

Explanations

words related to employment, such as employ, employed, and employs, and numerical indicators like the number of people employed

oai_token-act-pair · gpt-3.5-turbo

terms related to employment and workforce

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Patriarch

-0.67

etsk

-0.66

opy

-0.65

rx

-0.64

DragonMagazine

-0.61

point

-0.60

points

-0.59

 Strait

-0.59

omial

-0.58

ocol

-0.57

POSITIVE LOGITS

rious

0.84

eers

0.80

 laborers

0.76

ments

0.75

 interns

0.72

 workers

0.72

 secretaries

0.68

 managers

0.68

ancies

0.68

ment

0.68

Activations Density 0.027%

words related to employment, such as employ, employed, and employs, and numerical indicators like the number of people employed

terms related to employment and workforce

No Comments

No Known Activations