INDEX

Explanations

names of websites or acronyms starting with 'NS' or 'FW'

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

ously

-0.72

icago

-0.72

fare

-0.67

è¦ļéĨĴ

-0.65

thumbnails

-0.63

iated

-0.63

 Hundred

-0.62

iences

-0.62

âĸ¬âĸ¬

-0.62

ienced

-0.61

POSITIVE LOGITS

FW

1.47

fw

1.08

ERC

0.92

HT

0.91

erv

0.89

daq

0.87

erve

0.86

erves

0.83

zsche

0.83

FG

0.81

Activations Density 0.021%

names of websites or acronyms starting with 'NS' or 'FW'

No Comments

No Known Activations