INDEX

Explanations

phrases related to gratitude and appreciation

oai_token-act-pair · gpt-3.5-turbo

references to a collective or generalized group

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

 Kamp

-0.76

illac

-0.66

abwe

-0.62

yip

-0.62

pu

-0.59

potion

-0.59

fman

-0.59

 Presence

-0.58

kamp

-0.57

 Peninsula

-0.57

POSITIVE LOGITS

 sorts

1.19

 kinds

1.17

ocating

1.16

igators

0.95

iances

0.94

usions

0.94

igator

0.89

uv

0.84

ogene

0.83

uding

0.83

Activations Density 0.114%

phrases related to gratitude and appreciation

references to a collective or generalized group

No Comments

No Known Activations