INDEX

Explanations

contractions containing an apostrophe followed by the word 't'

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

CI

-0.66

 Appears

-0.66

imity

-0.63

amer

-0.62

ItemImage

-0.62

soDeliveryDate

-0.61

 bount

-0.61

senal

-0.58

Franc

-0.58

estate

-0.58

POSITIVE LOGITS

 afford

1.39

 imagine

0.98

 cope

0.97

 reproduce

0.95

 tolerate

0.94

 muster

0.93

 feas

0.92

 conceive

0.90

get

0.90

 handle

0.89

Activations Density 0.036%

contractions containing an apostrophe followed by the word 't'

No Comments

No Known Activations