INDEX

Explanations

phrases related to recurring occurrences or actions

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

riad

-0.73

ghai

-0.68

heng

-0.67

unning

-0.64

 likened

-0.60

 beforehand

-0.59

 Pipeline

-0.59

 aloud

-0.59

acebook

-0.58

oi

-0.56

POSITIVE LOGITS

 vengeance

0.93

 refreshed

0.92

 haunt

0.92

 rejuven

0.86

 again

0.85

 renewed

0.83

again

0.80

Reloaded

0.80

²¾

0.78

 reinforcements

0.76

Activations Density 0.265%

phrases related to recurring occurrences or actions

No Comments

No Known Activations