INDEX

Explanations

words related to gaming like "attack," "damage," "weapon," "ability," and "special."

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

isol

-0.24

 dstg

-0.22

cipled

-0.22

 Nadu

-0.21

arton

-0.21

é¾įåĸļå£«

-0.21

quickShipAvailable

-0.21

Wem

-0.20

mate

-0.20

istically

-0.20

POSITIVE LOGITS

#$

0.26

ERC

0.23

OE

0.22

 dodged

0.22

è¦ļéĨĴ

0.21

URI

0.20

[+]

0.20

 misses

0.20

Attack

0.20

/+

0.20

Activations Density 11.228%

words related to gaming like "attack," "damage," "weapon," "ability," and "special."

No Comments

No Known Activations