INDEX

Explanations

phrases related to conditions or situations that may or may not happen, often starting with "if that"

oai_token-act-pair · gpt-3.5-turbo

the word "that" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

Merit

-0.71

ãĥ©ãĥ³

-0.69

ITNESS

-0.67

ciples

-0.66

 Flavoring

-0.64

rior

-0.64

 Prepar

-0.64

IDES

-0.64

 Occupations

-0.63

ATURES

-0.63

POSITIVE LOGITS

 happens

1.01

 pesky

0.98

 happened

0.98

cher

0.91

 entails

0.89

 mattered

0.89

 occurs

0.87

 kind

0.86

 translates

0.82

 occurred

0.80

Activations Density 0.115%

phrases related to conditions or situations that may or may not happen, often starting with "if that"

the word "that" in various contexts

No Comments

No Known Activations