INDEX

Explanations

phrases related to progression or continuation in a story or narrative

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

Negative Logits

catentry

-0.74

 warr

-0.63

Cth

-0.62

oreal

-0.62

 convol

-0.60

anish

-0.58

tein

-0.58

ricted

-0.58

arov

-0.58

addons

-0.58

POSITIVE LOGITS

!:

0.93

âĶĢ

0.87

0.86

!.

0.81

!!!

0.81

 ðŁĳ

0.77

!!!!

0.76

!!

0.76

0.75

:#

0.74

Activations Density 0.445%

phrases related to progression or continuation in a story or narrative

No Known Activations