INDEX

Explanations

various phrases indicating alternatives or differing perspectives

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

urated

-0.84

usters

-0.81

osponsors

-0.74

gur

-0.68

oru

-0.65

ardy

-0.64

andise

-0.62

ilings

-0.62

mented

-0.62

ĸļ

-0.61

POSITIVE LOGITS

 sucks

0.71

 else

0.70

 though

0.68

 imaginable

0.66

 depends

0.65

soever

0.64

0.62

 however

0.61

 wors

0.60

we

0.60

Activations Density 0.008%

various phrases indicating alternatives or differing perspectives

No Comments

No Known Activations