INDEX

Explanations

words related to social interactions or relationships among individuals

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

 Hacker

-0.63

 Rite

-0.61

 Jihad

-0.60

($)

-0.59

 Tsukuyomi

-0.59

 countdown

-0.58

âĢ¢âĢ¢

-0.58

 Traff

-0.58

VALUE

-0.58

Spa

-0.58

POSITIVE LOGITS

hing

1.10

hed

0.98

orial

0.95

ured

0.92

oming

0.91

ress

0.89

nered

0.89

ruck

0.89

een

0.87

eenth

0.87

Activations Density 0.093%

words related to social interactions or relationships among individuals

No Comments

No Known Activations