INDEX

Explanations

places or locations

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

userc

-0.70

 Supplemental

-0.62

afety

-0.62

âĹ

-0.61

 stub

-0.61

©¶æ

-0.59

 NETWORK

-0.59

hower

-0.59

ocument

-0.58

 pretext

-0.57

POSITIVE LOGITS

agos

0.92

forth

0.78

oha

0.75

gow

0.73

reath

0.71

rant

0.70

furt

0.68

asks

0.68

ted

0.68

stad

0.67

Activations Density 0.089%

places or locations

No Comments

No Known Activations