INDEX

Explanations

references to economic aspects, especially related to unemployment benefits and rates

oai_token-act-pair · gpt-3.5-turbo

terms and phrases related to unemployment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

roman

-0.76

 orally

-0.72

gran

-0.70

sonian

-0.70

orses

-0.70

ILE

-0.68

wcsstore

-0.68

Zup

-0.67

yss

-0.66

afort

-0.66

POSITIVE LOGITS

 Unemployment

1.04

 unemployment

0.93

 unemploy

0.85

employment

0.82

 compensation

0.79

 unemployed

0.78

 Compensation

0.77

 contag

0.75

©¶æ¥µ

0.75

uates

0.74

Activations Density 0.014%

references to economic aspects, especially related to unemployment benefits and rates

terms and phrases related to unemployment

No Comments

No Known Activations