INDEX

Explanations

technical jargon related to cybersecurity and defense mechanisms

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

 colourful

-0.87

 favourites

-0.85

 humour

-0.81

Origin

-0.72

ĸļ

-0.72

 enqu

-0.70

 english

-0.69

 rumours

-0.67

iquette

-0.67

 Orient

-0.67

POSITIVE LOGITS

Assuming

1.06

 hypothetical

1.03

assuming

1.02

CBO

1.02

 Suppose

0.96

ivably

0.96

 scenario

0.95

would

0.94

 theoretically

0.94

Activations Density 0.454%

technical jargon related to cybersecurity and defense mechanisms

No Comments

No Known Activations