INDEX

Explanations

words related to technology and technical instructions

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

BuyableInstoreAndOnline

-0.71

DragonMagazine

-0.67

 Pebble

-0.65

uations

-0.63

ONES

-0.63

 GEAR

-0.62

 Nest

-0.62

 Sammy

-0.58

 buckle

-0.58

 simulator

-0.57

POSITIVE LOGITS

versible

1.15

porter

1.14

venge

1.14

ceiver

1.14

gex

1.12

ggie

1.12

asons

1.11

becca

1.10

ceive

1.09

levant

1.08

Activations Density 0.333%

words related to technology and technical instructions

No Comments

No Known Activations