INDEX

Explanations

phrases comparing two entities or situations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the."

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

 besides

-0.84

 furthermore

-0.76

 namely

-0.73

âĢł

-0.72

 followed

-0.70

elaide

-0.69

gpu

-0.69

istries

-0.68

 alongside

-0.68

âĢķ

-0.68

POSITIVE LOGITS

 slightest

1.24

 usual

1.20

 aforementioned

1.13

 smallest

1.12

 entirety

1.11

 rest

1.06

 proverbial

1.01

 latter

1.00

 prevailing

1.00

 same

0.99

Activations Density 0.400%

phrases comparing two entities or situations

instances of the word "the."

No Comments

No Known Activations