INDEX

Explanations

adjectives emphasizing extreme negative judgement

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anners

-0.92

actionDate

-0.88

azine

-0.78

roxy

-0.74

 Cells

-0.74

utics

-0.73

acters

-0.73

ernels

-0.71

azel

-0.71

poon

-0.70

POSITIVE LOGITS

 contradiction

1.17

 disregard

1.12

 impossibility

1.10

 absence

1.01

 absurdity

1.00

 ignorance

0.99

 necessity

0.97

 lack

0.97

 inability

0.96

 disbelief

0.96

Activations Density 0.122%

adjectives emphasizing extreme negative judgement

No Comments

No Known Activations