INDEX

Explanations

references to statistics or specific numerical values

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Stat" and its variations related to statistics or data reporting

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

intosh

-0.73

ONSORED

-0.72

 possibilities

-0.71

 traveler

-0.71

ï¸

-0.69

 traveller

-0.69

wash

-0.66

gie

-0.66

 Manz

-0.65

âĢ¢âĢ¢

-0.63

POSITIVE LOGITS

utes

1.06

utory

1.05

ute

1.05

ements

0.99

UTE

0.96

uting

0.91

idences

0.88

istically

0.87

istical

0.85

emic

0.84

Activations Density 0.035%

references to statistics or specific numerical values

occurrences of the word "Stat" and its variations related to statistics or data reporting

No Comments

No Known Activations