INDEX

Explanations

phrases related to actions that need to be taken

oai_token-act-pair · gpt-3.5-turbo

phrases indicating intentions or objectives

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

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

MpServer

-0.72

 respects

-0.64

*/(

-0.62

 Ambro

-0.62

 readings

-0.60

 IMAGES

-0.60

itudinal

-0.60

bats

-0.60

interstitial

-0.60

 impressions

-0.60

POSITIVE LOGITS

 spare

1.35

 contend

1.18

 choose

1.13

 conquer

1.06

 explore

1.03

 prove

0.98

 dispose

0.94

 spend

0.94

 negotiate

0.92

 expend

0.92

Activations Density 0.071%

phrases related to actions that need to be taken

phrases indicating intentions or objectives

No Comments

No Known Activations