INDEX

Explanations

phrases describing various things or people collectively

oai_token-act-pair · gpt-3.5-turbo

references to collective groups or universal themes

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

aminer

-0.78

yip

-0.72

SHIP

-0.70

bal

-0.69

 Kamp

-0.63

zag

-0.62

IDS

-0.60

oute

-0.60

sein

-0.59

bledon

-0.58

POSITIVE LOGITS

ocating

1.39

usions

1.19

ocated

1.15

uding

1.12

usion

1.12

igator

1.11

igators

1.09

udes

1.08

owing

1.04

 kinds

1.03

Activations Density 0.120%

phrases describing various things or people collectively

references to collective groups or universal themes

No Comments

No Known Activations