INDEX

Explanations

comparisons using the word "than"

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

ahime

-0.96

nown

-0.87

 Offline

-0.73

abin

-0.68

byter

-0.66

enser

-0.65

iliary

-0.64

devices

-0.63

anta

-0.63

ibal

-0.62

POSITIVE LOGITS

 pure

0.69

 guesses

0.67

anke

0.67

 oneself

0.65

 suffice

0.64

Jav

0.63

 superficial

0.63

 wish

0.62

 Revenge

0.61

 mediocre

0.61

Activations Density 0.155%

comparisons using the word "than"

No Comments

No Known Activations