INDEX

Explanations

references to differences in laws or regulations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

soever

-0.86

netflix

-0.81

din

-0.79

Ĭ

-0.77

linger

-0.75

¤

-0.72

LV

-0.71

ãĤ®

-0.70

NPR

-0.69

Nov

-0.69

POSITIVE LOGITS

 regards

1.13

 terms

1.06

 outcomes

0.98

 responsiveness

0.97

 relation

0.97

 favor

0.97

 regard

0.94

 favour

0.93

 styles

0.92

 degrees

0.92

Activations Density 0.046%

references to differences in laws or regulations

No Comments

No Known Activations