INDEX

Explanations

words related to effectiveness or efficiency

oai_token-act-pair · gpt-3.5-turbo

instances of the word "effective" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

forest

-0.70

Hop

-0.65

Boh

-0.65

umble

-0.64

IGH

-0.64

 Alive

-0.63

pper

-0.62

Fet

-0.60

 Yose

-0.59

 Fever

-0.59

POSITIVE LOGITS

 altru

1.03

iveness

0.88

ative

0.80

 deterrent

0.80

cheat

0.78

ively

0.75

rations

0.75

abulary

0.74

 contraception

0.73

acies

0.73

Activations Density 0.040%

words related to effectiveness or efficiency

instances of the word "effective" and its variations in different contexts

No Comments

No Known Activations