INDEX

Explanations

details related to military and law enforcement operations

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hex

-0.68

lahoma

-0.66

olulu

-0.64

gew

-0.63

idae

-0.63

acity

-0.63

ks

-0.62

gebra

-0.62

perties

-0.61

ghai

-0.61

POSITIVE LOGITS

ING

1.71

 EDITION

1.62

WOR

1.61

LY

1.59

ION

1.59

HEAD

1.59

ERS

1.58

MENT

1.58

MAN

1.58

ER

1.57

Activations Density 0.828%

details related to military and law enforcement operations

No Comments

No Known Activations