INDEX

Explanations

positively charged descriptive phrases regarding events or objects

oai_token-act-pair · gpt-3.5-turbo

phrases expressing superlative comparisons or preferences

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

stairs

-0.69

edom

-0.68

ropy

-0.66

naires

-0.63

abbage

-0.63

giene

-0.62

igham

-0.62

monary

-0.60

ities

-0.59

aine

-0.59

POSITIVE LOGITS

 than

0.92

 testament

0.78

 encaps

0.74

nor

0.73

 Than

0.72

âĢ¢âĢ¢âĢ¢âĢ¢

0.71

 exempl

0.69

GGGG

0.68

 embodies

0.67

 deserving

0.65

Activations Density 0.096%

positively charged descriptive phrases regarding events or objects

phrases expressing superlative comparisons or preferences

No Comments

No Known Activations