INDEX

Explanations

instances of the word "worked" and its variations in the context of employment or job-related experiences

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

emonic

-0.79

 constitu

-0.76

xual

-0.69

angular

-0.65

ĪĴ

-0.65

ople

-0.65

 Bubble

-0.64

 Augustus

-0.62

neum

-0.62

anus

-0.61

POSITIVE LOGITS

 tirelessly

1.09

 ethic

1.00

hops

1.00

 diligently

0.98

 overtime

0.93

bench

0.91

 collabor

0.90

 undercover

0.85

aday

0.84

station

0.81

Activations Density 0.058%

instances of the word "worked" and its variations in the context of employment or job-related experiences

No Comments

No Known Activations