INDEX

Explanations

the word "out" at different intensities

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 heartbeat

-0.60

 Bulg

-0.59

 capt

-0.58

 Yamato

-0.58

 Brach

-0.57

Rao

-0.55

 synonymous

-0.54

 complementary

-0.54

Nem

-0.53

eries

-0.52

POSITIVE LOGITS

stretched

1.02

doors

0.99

door

0.98

fitted

0.95

smart

0.91

ta

0.91

bur

0.85

skirts

0.84

rage

0.82

posts

0.80

Activations Density 0.034%

the word "out" at different intensities

No Comments

No Known Activations