INDEX

Explanations

instances of the word "way" where it has high relevance to the content

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

usters

-1.01

uster

-0.90

ĸļ

-0.80

 livest

-0.79

oppable

-0.72

inately

-0.72

omore

-0.70

ifer

-0.69

è¦ļéĨĴ

-0.69

icio

-0.68

POSITIVE LOGITS

fare

1.24

ward

1.18

finding

1.16

point

1.05

forward

1.04

points

0.91

 forward

0.88

bill

0.84

finder

0.84

station

0.78

Activations Density 0.474%

instances of the word "way" where it has high relevance to the content

No Comments

No Known Activations