INDEX

Explanations

phrases related to decision-making and uncertainty

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

Catalog

-0.97

thal

-0.78

lator

-0.77

oxide

-0.76

atari

-0.76

oola

-0.75

oven

-0.72

rites

-0.70

axter

-0.69

odder

-0.69

POSITIVE LOGITS

soever

0.87

 anyone

0.85

 anybody

0.78

 they

0.78

 consciously

0.70

 there

0.69

 intentional

0.68

it

0.67

 respondents

0.67

 exactly

0.66

Activations Density 0.958%

phrases related to decision-making and uncertainty

No Comments

No Known Activations