INDEX

Explanations

words or phrases indicating negation or disproval

oai_token-act-pair · gpt-3.5-turbo

negations in statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

stakes

-0.73

å¥

-0.69

éĥ

-0.68

æ©

-0.67

iers

-0.67

LG

-0.65

 Inventory

-0.64

PDATE

-0.64

ãĤ¼

-0.64

WER

-0.62

POSITIVE LOGITS

icably

1.33

icable

1.17

 necessarily

1.15

epad

1.10

hin

1.08

 exactly

0.99

orious

0.97

eworthy

0.90

 quite

0.87

yet

0.86

Activations Density 0.079%

words or phrases indicating negation or disproval

negations in statements

No Comments

No Known Activations