INDEX

Explanations

phrases that express expectations or predictions

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

predictions or expectations related to various topics or events

oai_token-act-pair · gpt-3.5-turbo

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

tein

-0.76

formerly

-0.72

utterstock

-0.72

arta

-0.71

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

-0.68

 Tycoon

-0.68

abase

-0.67

olate

-0.64

ccess

-0.63

IFT

-0.63

POSITIVE LOGITS

 surprises

1.21

 backlash

1.07

 outcry

1.03

 announcements

1.01

 uptick

1.00

 uproar

0.98

 influx

0.95

 resurgence

0.94

 flurry

0.92

 upheaval

0.90

Activations Density 0.370%

phrases that express expectations or predictions

predictions or expectations related to various topics or events

No Comments

No Known Activations