INDEX

Explanations

references to executive orders

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

INGTON

-0.88

bane

-0.82

Else

-0.81

ulia

-0.81

MQ

-0.78

anners

-0.77

selves

-0.77

CTV

-0.75

schild

-0.73

debian

-0.72

POSITIVE LOGITS

 branch

0.99

 privilege

0.98

 amnesty

0.96

 memorandum

0.94

 summary

0.94

 authority

0.92

ctive

0.92

 order

0.89

 actions

0.89

 orders

0.87

Activations Density 0.008%

references to executive orders

No Comments

No Known Activations