INDEX

Explanations

the word "zero" paired with numerical values suggesting a scale or range

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a lack or absence of something, particularly "zero tolerance" concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

rough

-0.80

ammad

-0.72

nikov

-0.71

andise

-0.71

tha

-0.69

Downloadha

-0.68

regor

-0.68

tub

-0.68

akov

-0.67

trl

-0.66

POSITIVE LOGITS

MQ

0.95

 tolerance

0.91

 percent

0.80

 gravity

0.80

 Hedge

0.79

pointer

0.75

 Gravity

0.74

 amount

0.73

lessly

0.70

xd

0.69

Activations Density 0.031%

the word "zero" paired with numerical values suggesting a scale or range

phrases that indicate a lack or absence of something, particularly "zero tolerance" concepts

No Comments

No Known Activations