INDEX

Explanations

phrases related to instructions or directives

oai_token-act-pair · gpt-3.5-turbo

terms related to organizational structures and roles

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

entimes

-0.66

uristic

-0.63

 sometimes

-0.62

ozy

-0.61

 proverb

-0.61

slaught

-0.60

hua

-0.59

whel

-0.59

eka

-0.56

erick

-0.56

POSITIVE LOGITS

 except

1.67

 imaginable

1.27

except

1.21

 alike

1.00

 whatsoever

0.90

 irrespective

0.88

 including

0.86

 together

0.85

 sexes

0.85

 Including

0.84

Activations Density 0.304%

phrases related to instructions or directives

terms related to organizational structures and roles

No Comments

No Known Activations