INDEX

Explanations

phrases indicating superlatives or extremes

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ERAL

-0.73

ose

-0.69

icio

-0.68

avascript

-0.68

ice

-0.66

rar

-0.66

resy

-0.64

 Bullets

-0.64

ensing

-0.62

rapnel

-0.62

POSITIVE LOGITS

 recorded

1.13

 conceived

0.95

 produced

0.95

 seen

0.93

 achieved

0.92

 devised

0.90

 attained

0.85

 bestowed

0.84

 undertaken

0.83

 assembled

0.80

Activations Density 0.025%

phrases indicating superlatives or extremes

No Comments

No Known Activations