INDEX

Explanations

positive or significant news

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

senal

-0.80

urat

-0.74

sf

-0.68

NetMessage

-0.67

ONSORED

-0.66

dq

-0.65

ĸļ

-0.63

ngth

-0.62

regor

-0.61

ivated

-0.60

POSITIVE LOGITS

Catholic

0.78

adays

0.76

vale

0.75

 outweigh

0.75

worthy

0.72

night

0.72

:,

0.72

tainment

0.72

 Roundup

0.71

peak

0.71

Activations Density 0.026%

positive or significant news

No Comments

No Known Activations