INDEX

Explanations

words and phrases related to conflicts and disputes

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_fs6144-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_6144

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

6,144

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

icrobial

-0.73

eele

-0.70

vation

-0.69

urances

-0.67

undai

-0.65

prints

-0.65

ola

-0.64

elta

-0.64

-0.63

ECD

-0.63

POSITIVE LOGITS

 raging

1.16

 between

1.14

 flared

1.10

 raged

1.08

 ensued

0.96

 waged

0.95

 pits

0.93

 rage

0.92

 involving

0.92

 erupted

0.90

Activations Density 0.483%

words and phrases related to conflicts and disputes

No Comments

No Known Activations