INDEX

Explanations

adverbs that enhance the description of actions or states

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

ammy

-0.84

Ki

-0.69

anty

-0.69

IRO

-0.68

 Dwell

-0.66

 DRAG

-0.64

 Redemption

-0.62

Quantity

-0.58

yg

-0.57

RIP

-0.57

POSITIVE LOGITS

aneously

0.82

ãĢĤ

0.80

 enough

0.77

 afterwards

0.76

 before

0.76

 lest

0.74

0.73

 whilst

0.72

itably

0.70

 afterward

0.70

Activations Density 0.133%

adverbs that enhance the description of actions or states

No Comments

No Known Activations