INDEX

Explanations

phrases related to academic and legal contexts

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

ersed

-0.74

otin

-0.72

bender

-0.69

witch

-0.69

escription

-0.68

trak

-0.68

nesty

-0.68

urations

-0.66

aphael

-0.66

verb

-0.65

POSITIVE LOGITS

own

1.75

OWN

1.02

 liking

1.01

 repertoire

1.00

 respective

0.99

Own

0.99

 arsenal

0.98

 fingertips

0.94

 pocket

0.87

 portfolio

0.86

Activations Density 0.177%

phrases related to academic and legal contexts

No Comments

No Known Activations