INDEX

Explanations

the word "there" followed by various contextual words and phrases

oai_token-act-pair · gpt-3.5-turbo

phrases indicating existence or presence

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ONSORED

-0.67

ointed

-0.61

shoot

-0.60

uously

-0.60

 wheelchair

-0.60

 externalToEVAOnly

-0.57

HI

-0.56

leeve

-0.55

Ã©e

-0.55

EA

-0.53

POSITIVE LOGITS

abouts

1.35

upon

0.97

 exists

0.88

 aren

0.86

'll

0.86

isn

0.85

are

0.84

fore

0.81

 happens

0.81

ARE

0.80

Activations Density 0.122%

the word "there" followed by various contextual words and phrases

phrases indicating existence or presence

No Comments

No Known Activations