INDEX

Explanations

words related to the opening of events or establishments

oai_token-act-pair · gpt-3.5-turbo

instances of the word "opened."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

iable

-0.66

orum

-0.66

grade

-0.63

bil

-0.63

ivable

-0.61

iane

-0.59

CTV

-0.58

cia

-0.58

dos

-0.58

âĨĳ

-0.58

POSITIVE LOGITS

 Doors

0.92

 doors

0.78

 wounds

0.75

 curtains

0.73

up

0.73

 parentheses

0.72

 hostilities

0.72

office

0.72

SSL

0.71

ĻĤ

0.71

Activations Density 0.034%

words related to the opening of events or establishments

instances of the word "opened."

No Comments

No Known Activations